This item is only available as the following downloads:
1 IS A FAST OR SLOW RATE OF WEIGHT LOSS ASSOCIATED WITH GREATER LONG TERM WEIGHT LOSS MAINTENANCE? THE EFFECTS OF PRESCRIBING MILD OR MODERATE CALORIE RESTRICTION GOALS By LISA MARIE NACKERS A DISSERTATION PRESENTED TO THE G RADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 201 2
2 201 2 Lisa Marie Nackers
3 To everyone who supported me as a graduate student over the past five years
4 ACKNOWLEDGEMENTS I would like to thank my mentor, Dr. Michael Perri for his support, guidance, and wisdom. I would also like to thank the members of my supervisory committee, Dr. David Janicke, Dr. Tracey Barnett, and Dr. Stephen Anton for their time and assistance. Additionally, I would like to thank my colleagues in the UF Weight Management Program lab for their constant encouragement and support on this project, especially Kathryn Ross for her assistance with the statistical analyses. Finally, I would like to thank my family for providing unwavering love and support and for encouraging me to continue striving toward the finish line.
5 T ABLE OF CONTENTS p age ACKNOWLEDGEMENTS ................................ ................................ ............................... 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIG URES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 11 Obesity ................................ ................................ ................................ .................... 14 Prevalence ................................ ................................ ................................ ....... 14 Health Conditions ................................ ................................ ............................. 14 Economic Factors ................................ ................................ ............................. 16 Lifestyle Intervention for Obesity ................................ ................................ ............. 17 The Maintenance Problem ................................ ................................ ...................... 20 Strategies for Im proving Weight Maintenance ................................ ........................ 21 Caloric Restriction and Rate of Weight Loss ................................ ........................... 23 Very Low Calorie Diets ................................ ................................ ..................... 23 Moderate Caloric Restriction ................................ ................................ ............ 24 Mild Caloric Restriction ................................ ................................ ..................... 26 Caloric Intake Prescription ................................ ................................ ...................... 29 Self Efficacy and Quality of Life ................................ ................................ ........ 30 Disinhibition, Restraint, Hunger, Depression, and Binge Eating ....................... 31 Improving Caloric Intake Adherence ................................ ................................ ....... 34 Self Monitoring ................................ ................................ ................................ 34 Partial Meal Replacements ................................ ................................ ............... 35 Portion Controlled Meals ................................ ................................ .................. 36 Structured Meal Plans ................................ ................................ ...................... 37 Cognitive Be havioral Strategies ................................ ................................ ....... 37 Specific Aims and Hypotheses ................................ ................................ ............... 38 2 MATERIALS AND METHODS ................................ ................................ ................ 40 Research Methods and Procedures ................................ ................................ ....... 40 Participants ................................ ................................ ................................ ....... 40 Procedure ................................ ................................ ................................ ......... 41 Measures ................................ ................................ ................................ .......... 44
6 Statistical Analyses ................................ ................................ ................................ 47 Primary Aim ................................ ................................ ................................ ...... 47 Secondary Aim ................................ ................................ ................................ 48 Tertiary and Exploratory Aims ................................ ................................ .......... 48 3 RESULTS ................................ ................................ ................................ ............... 50 Participants ................................ ................................ ................................ ............. 50 Primary Aim ................................ ................................ ................................ ............ 51 Weight Change Outcomes ................................ ................................ ............... 51 Weight Change at Month 6 ................................ ................................ .............. 52 Weight Change at Month 12 ................................ ................................ ............ 52 Secondary Aim ................................ ................................ ................................ ....... 53 Tertiary and Exploratory Aims ................................ ................................ ................. 55 Restraint, Disinhibition, Hunger ................................ ................................ ........ 55 Depression ................................ ................................ ................................ ....... 56 Self Efficacy ................................ ................................ ................................ ..... 57 Mediation of Binge Eating ................................ ................................ ................. 57 Quality of Life ................................ ................................ ................................ ... 58 Weight Regain ................................ ................................ ................................ 59 4 DISCUSSION ................................ ................................ ................................ ......... 71 Primary Aim ................................ ................................ ................................ ............ 71 Secondary Aims ................................ ................................ ................................ ...... 77 Tertiary and Exploratory Aims ................................ ................................ ................. 78 Limitations ................................ ................................ ................................ ............... 85 Clinical Implications ................................ ................................ ................................ 86 APPENDI X A SAMPLE METHODS TO MEET CALORIE INTAKE GOAL ................................ .... 91 B GROUP PERFERENCE QUESTI ONNAIRE ................................ ........................... 95 LIST OF REFERENCES ................................ ................................ ............................... 96 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 111
7 LIST OF TABLES Table Page 3 1 Baseline values according to caloric prescription ................................ ............... 63 3 2 Weight changes from Month 0 to Months 6 and 12 according to caloric prescription ................................ ................................ ................................ ......... 63 3 3 Attendance rates for Phase I and Phase II according to caloric prescription ...... 64 3 4 Weight changes according to matched and unmatched calor ic prescription preference and assignment ................................ ................................ ................ 64 3 5 Percent adherence categorization for Phase I according to caloric prescription ................................ ................................ ................................ ......... 64 3 6 Restraint, disinhibition, hunger, self efficacy, depression, and binge eating at Months 0, 6, and 12 according to caloric prescription ................................ ........ 65 3 7 Quality of life variables from the SF 3 6 at Months 0, 6, and 12 according to caloric prescription ................................ ................................ .............................. 66 3 8 Month 0 study variables according to weight regain from Months 6 to 12 .......... 67 3 9 Weight regain by Month 0 study variables according to treatment condition ...... 67 3 10 Intervention variables according to weight regain from Months 6 to 12 .............. 68 3 11 Month 6 and 12 study variables according to weight regain from Months 6 to 12 ................................ ................................ ................................ ....................... 68
8 LIST OF FIGURES Figure Page 3 1 CONSORT diagram of participation rates during the 12 month program. .......... 69 3 2 Adjusted weight changes a ccording to treatment condition ................................ 7 0
9 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy IS A FAST OR SLOW RATE OF WEIGHT LOSS ASSOCIATED WITH GREATER LONG TERM WEIGHT LOSS MAINTENANCE? THE EFFECTS OF PRESCRIBING MILD OR MODERATE CALORIE RESTRICTION GOALS By Lisa Marie Nackers August 2012 Chair: Michael G. Perri Major: Psychology Lifestyle intervention s for obesity typically result in 7 to 10% reduction in body weight; however, these beneficial losses are rarely maintained. Controversy exists regarding whether small or large changes in caloric intake result in greater long term weight reduction. This study examine d the effects of mild versus moderate caloric restriction on weight lo ss. Additional ly, the study assessed whether participants adhered to their caloric prescription Participants included 125 obese women (mean BMI = 37.9 + 3.9 kg/m 2 ; mean age = 52.0 + 10.8 yr) who were assigned either a 1,0 00 or 1,5 00 kcal/day goal Resu lts indicated that participants prescribed the 1,000 kcal/day goal lost more weight at Month 6 compared to those prescribed the 1,500 kcal/day goal ( 10.8 + 6.8 kg versus 6.3 + 6.8 kg, respectively, p = .04 ). From Month s 6 12 the 1,0 0 0 kcal/day condi tion regained more weight than the 1,500 kcal/day condition so that weight change at Month 12 was not significantly different ( 8.6 + 8.4 kg versus 5.8 + 8.4 kg, respectively, p = .212); however, 61 % of participants in the 1,000 kcal/day condition compare d to 42 % in the 1,500 kcal/day condition achieved a weight loss > 5%. T he majority of participants in both condition s adhered to their caloric
10 prescriptions with the 1,000 kcal/day and 1,500 kcal/day participant s consuming sig nificantly different amounts (1, 164 + 170 versus 1,518 + 222 kcal/day respectively, p < .001) Treatment condition moderate d the association between baseline caloric intake and weight regain ; participants reporting higher caloric intake at baseline who were assigned the 1,000 kcal/ day goal regained significan tly more weight from Months 6 12 than those consum ing high er baseline calories who were assigned the 1,500 kcal/day goal. F indings suggest that prescribing a 1,000 kcal/day goal results i n greater short term weight loss than a 1,500 kcal/day goal ; however, larger caloric restriction may increase susceptibility to weight regain especially for p articipants who consum e greater amounts of calories at baseline While power was in adequate to detect a significant long term weight l oss difference, results indicate that prescribing a 1,000 rather than 1,500 kcal/day goal increase s the likelihood of achieving a 5% weight reduction.
11 CHAPTER 1 INTRODUCTION With its dramatic increase in prevalence and direct link to f ive of the ten leading causes of death in the United States obesity (defined as a Body Mass Index, BMI, > 30 kg/m 2 ) has become one of the most significant public health epidemics (NHLBI, 1998; Flegal, Carroll, Ogden, & Curtin, 2010 ). Considerable researc h efforts have been placed on identifying interventions that result in a body weight reduction of a sufficient magnitude to improve health, f ound to be greater than 5 % ( IOM 1995 ) Lifestyle interventions for obesity, consisting of 1 6 to 2 6 weekly group s essions, typically result in weight losses of 7 to 10% and reduced health related risk factors (Jeffery et al., 2000; Perri & Fuller, 1995; Vetter, Faulconbridge, Webb, & Wadden, 2010; Wadden, Butryn, & Wilson, 2007) ; however, long term weight maintenance is rarely achieved (Perri, 1998; Wadden et al., 200 7 ) Research has suggested that early weight loss may serve as a predictor of long term success in weight management; however, controversy exists regarding an ideal amount of initial weight reduction (Ca rels Cacciapaglia, Douglass, 2003; Fogelholm, Kukkonen Harjula, & Oja, 1999; Jeffery, Wing, & Mayer 1998; Lutes et al., 2008; Sbrocco Nedegaard, Stone, & Lewis, 1999; Wadden et al., 1992). Recent initiatives promote the utilization of a small changes approach in which small presumably sustainable changes in caloric intake and physical activity may serve as an effective strategy for achieving and maintaining slow, yet steady reductions in body weight (Hill, 2009; Hill, Wyatt, Reed & P eters, 2003) Adopting this design in the context of lifestyle interventions may be effective for continued weight loss, reduced risk for weight regain, and successful long term weight loss maintenance ( Lutes, et al.,
12 2008; Sbrocco et al., 1999 ) Convers ely, recent reviews and studies have coupled larger initial weight loss during lifestyle intervention with greater long term body weight reductions (Astrup & R ssner, 2000; Carels et al., 2003; Elfhag & R ssner, 2005; Elfhag & Rssner, 2010; Jeffery et al. 1998; Nackers, Ross, & Perri, 2010 ; Wadden, et al., 1992). However, i n some cases, l arger initial weight loss has been associated with greater long term weight regain (Jeffery et al., 1998; Sbrocco et al., 1999; Wadden, Foster, & Letizia, 1994), suggest ing larger early weight reduction may serve as a risk factor for later weight regain (McGuire, Wing, Klem, Lang, & Hill, 1999; Weiss et al., 2007; Wing & Hill, 2001), and thus bringing to question the ability of participants to maintain large changes in di et and exercise following lifestyle intervention However, with the use of continuous contact or extended care programs, maintenance of significant weight losses may be improved (Nackers, et al. 2010 ; NHLBI, 1998; Perri et al., 2008). To our knowledge, only two studies have attempted to prospectively p rescribe differing levels of caloric intake within the recommended range exceeding 1,000 kcal/day (Das et al., 2009; Toplak et al., 2005). While the majority of participants in both studies achieved weigh t losses of 5 to 10%, neither of these studies found differences in long term weight loss between participants who either mildly or moderately restricted caloric intake. Furthermore, when assessing caloric intake, both studies found actual calories consum ed were similar betw een condition s, with those who received prescriptions for mild caloric restriction consuming fewer calories than prescribed and those assigned to moderate caloric restriction consuming greater amounts of calories than prescribed (Das et al., 2009; Toplak et al., 2005). However, t he se studies were not
13 performed in the context of a lifestyle intervention with weekly group meetings that emphasized behavioral principles to attain energy goals. In addition, it is unclear whether prescribin g restrictive caloric goals prospectively res u lts in adverse dietary events. Research on restrained eaters, or chronic dieters who restrict their intake of food with the goal of achieving or maintaining a lower weight (Heatherton, Polivy, & Herman, 1991), indicates that dietary restraint does not necessarily lead to successful weight loss due to vacillation between bouts of intense caloric restriction and disinhibition (i.e. a breakdown in the conscious control of food intake) (Heatherton, Herman, Polivy, King, & McGree, 1988). In fact, restrained eaters tend to binge eat (i.e. consume large quantities of food in short periods of time) when self control is disrupted by critical events or disinhibitors (Marcus, Wing, & Lamparski, 1985; Ruderman, 1986). Bin ge eating, in turn, has been labeled a risk factor for increased body weight o ver time ( Klesges, Isbell, & Klesges, 1992 ). Notably, r eductions in disinhibition ha ve consistently been linked to greater success in weight loss and weight maintenance (Niemeie r, Phelan, Fava, & Wing, 2007; McGuire et al. 1999). It remains questionable then whether prospectively prescribing caloric intake goals that require greater dietary restraint will increase periods of disinhibition and adversely impact the extent and sev erity of binge eating than goals requiring lower dietary restraint Therefore, in sum it is unknown whether prescribing goals of either mild caloric reduct ion so as to achieve a gradual, steady weight loss, or moderate caloric reduction so as to achieve a fast, large initial weight loss will result in greater long term weight reduction for participants partaking in a l ifestyle intervention. In addition, it is unknown
14 whether participants assigned to these energy restriction goals adhere to their prescribe d levels of caloric intake and whether they experience altered levels of restraint or disinhibition or increases in binge eating episodes Th is dissertation sought to answer these questions utilizing a 24 week lifestyle intervention and a 6 month extended care program. Obesity Prevalence Obesity results from an imbalance of the energy equation where energy intake (i.e. caloric intake) exceeds energy expenditure (i.e. metabolic rate and physical activity) (Tataranni & Ravussin, 2002). Reports from the Nati onal Health and Nutrition Examination Survey (NHANES) illustrate that rates of obesity in adults rose from 12.8% in 1960 1962 (Flegal, Carroll, Kuczmarski, & Johnson, 1998) to 33.8% in 2007 2008 ( Flegal et al., 20 10 ) with older adults, women those from l ow socio economic status and inhabitants of rural areas exhibiting higher rates of obesity (Chang & Lauderdale, 2005; F legal, Carroll, Ogden, & Johnson, 2002; Hedley et al., 2004; Flegal et al., 20 1 0; Patterson, Moo re, Probst, & Shinogle, 2004). Health C onditions Evidence directly links excess weight with an increased morbidity. Obesity is associated with increased risk for cardiovascular disease (Flegal, Graubard, Williamson, el, 2002), hypertension (Mokdad et al., 2003), hypercholesterolemia (Brown et al., 2000; Mokdad et al., 2003), type 2 diabetes (Flegal et al., 2007; Mokdad et al., 2003), asthma, arthritis (Mokdad et al., 2003), gallbladder disease (Field et al., 2001), sl eep apnea (Vgontzas et al., 1994), renal disease (Flegal et al., 2007), and breast, pancreatic, and prostate
15 cancers (Dumitrescu & Cotalra, 2005; Field et al., 2001; Freedland & Aronson, 2005). Obese individuals are also 2.8 times more likely to experienc e functional impairment (e.g. walking or lifting) than their normal weight counterparts (Alley & Chang, 2007). Obesity is also related to an increased mortality and decreased life expectancy. Conservative estimates suggest 111,000 excess deaths per year are attributable to obesity (Flegal, Graubard, Williamson, & Gail, 2005) while others approximate numbers of deaths as high as 400,000 (Mokdad, Marks, Stroup, & Gerberding, 2004). R esults from the Framingham Heart Study suggest that BMI at ages 30 to 49 predicts mortality in future decades, with five to seven years of life lost in obese individuals (Peeters et al., 2003). In addition to adverse physical consequences obesity negatively affects psychosocial outcomes S tigmatization against obese person s in employment, education, and healthcare have been well documented (Puhl & Brownell, 2001) and close relationship partners (i.e. friends, parents, and spouses) serve as the worst sources of stigmatization (Puhl, Moss Racusin, Schwartz, & Brownell, 2008 ) Obese individuals experience lower quality of life and higher levels of depression (Fontaine & Barofsky, 2001; Jia & Lubetkin, 2005; Wadden, Womble, Stunkard, & Anderson, 2002). Obese women are also at increased risk for suicidality (Carpenter, Hasin, Allison, & Faith, 2000). Research demonstrates that weight loss can reverse many of the adverse effects associated with obesity Even a modest reduction in body weight of > 5 % produce s beneficial effects on glycemic control, hypertension, and hyperlipidemi a (NHLBI, 1998). The Look AHEAD study is an ongoing multicenter randomized controlled trial that
16 compares lifestyle intervention to diabetes support and education on the incidence of cardiovascular disease morbidity and mortality in overweight or obese pe rsons with type 2 diabetes (The Look AHEAD Research Group, 2010). Across four years, results indicate that individuals who received the lifestyle intervention lost a mean 6% bodyweight and experienced reductions in levels of HbA1c systolic blood pressure diastolic blood pressure, triglyceride s and LDL cholester ol. These values we re significantly improved compared to those in the education condition who lost 0.9% body weight (The Look AHEAD Research Group, 2010). O ther large scale lifestyle interventio n trials have found similar health related benefits in participants who lost weight compared to control condition s who do not achieve significant weight losses (Diabetes Preventi on Program Research Group, 2002; Stevens et al., 2001; Tuomilehto et al., 2001 ). However, weight regain may also reverse these positive health outcomes and therefore, sustained weight loss remains an important goal (Moore et al., 2000; Stevens et al., 2001; Wadden, Anderson, & Foster, 1999). Economic Factors The hazardous health consequences of obesity are accompanied by a significant financial burden. In the United States during 2008 the healthcare costs associat ed with obesity totaled about $ 147 billion (Finkelstein, Trogdon, Cohen, et al., 2009). In the year 200 6 medical ex penditures for obese people was $1,429 (42%) greater than spending for someone of normal weight (Finkelstein et al., 2009). It has also been estimated that public sector Medicare and Medicaid spending, in the absence of obesity, would be 8.5% and 11.8% lo wer, respectively (Finkelstein et al., 2009).
17 Lifestyle Intervention for Obesity With the increased prevalence and adverse outcomes a ssociated with obesity comes a need for effective weig ht loss treatments. Lifestyle interventions represent the first l ine of professional treatment for obesity (NHLBI, 199 8 ). These programs teach participants to utilize cognitive behavioral strategies to modify eating and activity patterns to produce a negative energy balance necessary to lose weight ( Vetter et al., 2010 ). By applying the classical and operant learning theory to weight loss, participants learn to identify and modify the antecedents and consequences associated with their unhealthy eating patterns and low levels of physical activity and to alter their envi ronment to increase and develop habits that promote healthy behaviors ( Ferster, Nurnberger, & Levitt, 1962; Stuart, 1967; Vetter et al., 2010; Wing, 2002). According to the NHLBI (1998), decreasing caloric intake serves as the most important element for weight reduction and maintenance. Whereas one pound is equivalent to 3,500 kcal, reducing daily caloric intake by 500 to 1,000 kcal typically results in a weight loss of 0.5 to 0.9 kg (i.e. 1 to 2 lbs) per week (NHLBI, 1998). Because requirements for mai ntaining the energy balance is not often known, researchers assume that resting energy expenditure approximates 1 kcal/kg*h, and therefore often prescribe diets of 1,000 to 1,500 kcal/day for obese women to achieve the aforementioned weekly weight losses ( Brownell, 2000; Wing, 1998). This level of energy intake has been shown to be safe and effective in women averaging 90.7 kg (i.e. 200 lbs) who aspire to lose weight (Jakicic, et al., 2001). In addition to reduction of caloric intake, consumption of a b alanced diet has been associated with weight loss. A balanced deficit diet consists of caloric reduction while emphasizing consumption of all food groups and a variety of nutrients, (Melanson
18 & Dwyer, 2002). Lifestyle interventions typically encourage pa rticipants to adhere to a balanced deficit diet in which they reduce their caloric and fat intake and increase amounts of fruits, vegetables, and whole grains (Wing, 2002). For example, lifestyle intervention s, such as the Tr eatment of Obesity in Underser ved Rural Settings (TOURS) project, set goals for participants to reduce caloric intake by 500 1,000 kcal/day, and modification of total fat intake to <25 30%, saturated fat to 7%, and protein to 15% of total caloric intake (Perri et al., 2008). In additi on, participants were encouraged to increase fruit and vegetable consumption to at least 5 per day and whole grains to 3 or more servings per day (Perri et al., 2008). On the other half of the equation, increasing physical ac tivity to augment energy expend iture further creates a negative energy deficit. The m ost recent recommendations have suggested that overweight and obese adults participate in at least 150 minutes/week of moderate intensity physical activity (i.e. walking) to prevent weight gain, elicit modest reductions in body weight, and reduce risk factors for chronic disease (Donnelly, Blair, Jakicic, Manore, Rankin, & Smith, 2009). However, researchers have concluded that there exists a dose effect of physical activity where approximately 250 to 3 00 minutes/week (i.e. approximately 2,000 kcal/week) of moderate intensity physical activity produces greater weight loss and prevention of weight regain, suggesting overweight and obese adults progress to greater amounts of physical activity for larger be nefits (Donnelly et al., 2009). In addition, reducing sedentary time by increasing every day, lifestyle, activities is encouraged (NHLBI, 1998). While adding 30 60 minutes of physical activity three times per week to a program of caloric restriction has only been shown to increase weight loss modestly by
19 approximately 2 kg, regular physical activity plays an important role in maintenance of weight loss (Blair & Leermakers, 2002). In addition, cognitive behavioral strategies utilized in lifestyle intervent ions are consistent with the social cognitive theory (Wadden & Foster, 1992). This theory describes how personal factors, such as cognitions and emotions, and environmental ve a reciprocal influence on personal and environmental factors (Bandura, 1977, 1986). Therefore, lifestyle interventions target four key concepts of the social cognitive theory (Bandura, 1977, 1986; Wadden & Foster, 1992) First, they are used to increa se knowledge of health behaviors, such as diet and physical activity, and their effects on to perform positive behaviors and create positive outcome expectancies th rough promotion of successful eating and exercise experiences. Third, they increase ability to exert control over his or her behavior, cognitions, and environment through the use of goal setting, self monitoring of daily food intake and physical act ivity, self reinforcement, stimulus contro l, and cognitive restructuring And fourth, lifestyle interventions teach participant s how to overcome barriers to performing specific beha viors through problem solving. National guidelines purport that a combine d intervention utilizing the aforementioned cognitive behavioral techniques, a caloric restricted diet, and increased physical activity results in the most successful non surgical treatment for weight loss and weight maintenance (NHLBI, 1998). Lifestyle in terventions have been found to be more effective when offered in groups versus individual treatment regardless of preferred treatment modality
20 ( Renjilian, Perri, Nezu, McK elvey, Shermer, & Anton, 2001). Additional studies of treatment preference have ind icated that participants who are randomly matched to their preferred type of diet ary treatment or to their preferred weight maintenance treatment guidelines achieve no greater weight loss outcomes than those who are matched to their non preferred treatment options (Burke et al., 2008 ; Vogels & Westerterp Plantenga, 2005; Warziski, Sereika, Styn, Music, & Burke, 2008). Therefore, assigning participants to treatment of choice does not necessarily lead to improved adherence or greater weight maintenance. The Maintenance Problem Reviews of randomized trials suggest that comprehensive behavioral modification programs, which typically consist of weekly group treatment sessions for four to six months, result in a 7 to 10% reduction of initial body weight, or a mea n loss of 0.5 kg per week (Jeffery et al., 2000; Perri & Fuller, 1995; Wadden et al., 2007 ). Weight losses of this magnitude have been associated with reductions in adverse health conditions and risk factors associated with obesity ( NHLBI 1998). Unfortu nately, long term follow up evaluations indicate that traditional behavioral treatment induced weight losses are not well maintained (Perri, 1998; Wadden et al., 200 7 ). A review by Wadden et al. (2007 ) indicated that in the year following 20 to 30 weeks o f group lifestyle modification treatment, patients regain 30% to 35% of their lost weight. Weight regain typically slows after the first year, but after 18 months following program entry, participants maintain only about 50% of the initial lost weight (Je ffery et al., 2000). Five years after treatment, 50% or more of patients return to their initial body weight (Wadden, Sternberg, Letizia, Stunkard, & Foster, 1989). Another review found that only
21 13% to 22% of patients who initially lost > 5 kg maintaine d this weight loss at five year follow up (Wing & Hill, 2001). Thus, it appears that the pattern associated with behavioral treatment begins with successful initial weight loss followed by a reliable regain of lost weight (Jeffery et al., 2000). Strategie s for Improving Weight Maintenance Countless research efforts have sought to better understand the factors associated with the maintenance problem. R ecent reviews found that personal characteristics, such as self mo tivation, self efficacy, autonomy suffi cient coping strategies for handling stress, and overall psychological stability, as well as fewer previous weight loss attempts served as predictors of successful weight loss (Elfhag and Rssner, 2005; Teixeira, Going, Sardinha, & Lohman, 2005) In addit ion, results from the National Weight Control Registry, a group of over 3,000 people who have lost and maintained at least a 30 lb weight loss for one year, indicate that participating in high levels of physical activity, frequent self monitoring of body w eight and food intake, and eating a low calorie and low fat diet are associated with long term weight maintenance (Wing & Hill, 2001). Conversely, factors that may increase the likelihood of weight regain include decreased restraint with dietary intake, i ncreased hunger, disinhibited eating, binge eating, eating as a response to stress or negative emotions, passive reactions to problems, and a history of weight cycling (Elfhag & Rssner, 2005; McGuire et al., 1999). B ehavioral factors during weight manag ement programs, such as higher rates of attendance (Carels et al., 2003; Wadden et al., 1992) and adherence, as measured by self monitoring and recording daily dietary intake (Baker & Kirschenbaum, 1993;
22 Boutelle & Kirschenbaum, 1998), have also been corre lated with long term weight loss success. Perri and Corsica (2002) summarized the variety of strategies designed to improve lifestyle interventions and facilitate weight maintenance These include increasing the initial treatment leng th, engaging partici pants in extended care program s via in person or telephone sessions, utilizing exercise focused programs or personal trainers, incorporating peer support, providing food or incentives, and creating multicomponent treatment programs. Of these, one of the m ost successful strategies appears to be offering extended care programs in which participants are encouraged to practice maintenance of the behaviors they learned in the initial lifestyle intervention (Vetter et al., 2010) Perri and Corsica (2002) revie wed 13 studies that utilized extended care approaches spanning past six months of behavioral treatment. P articipants who received extended care maintained 96% of their weight losses compared to 67% in those who did not receive extended follow up care. Ex tended care programs allow continuous practice of strategies and behaviors learned during initial treatment at a time when rate o f weight loss begins to slow and participants struggle with maintaining lost weight (Perri & Corsica, 2002). Other reviews th at support extended care treatment for weight maintenance raise the question of whether additional factors, such as level of energy prescription for producing a specific rate of weight loss may result in augmented weight loss outcomes ( Astrup & Rssner, 2 000; Ayyad & Anderson, 2000). Current findings prove inconclusive and t herefore, r ecent debates center around the role of rate of weight loss on long term success in weight management.
23 Caloric Restriction and Rate of Weight Loss Very Low Calorie Diets Two decades ago, several studies assessed the effectiveness of severe dietary restriction, or very low calorie diets (VLCDs) consisting of an energy content of 400 800 kcal/day and often consumed in liquid formula (Wadden et al., 1994; Wadden & Stunkard, 1986; Wing, Blair, Marcus, Epstein, & Harvey, 1994; Wing et al., 1991). Because the energy deficit is often greater than that typically prescribed in conventional diets of 1,000 1,500 kcal/day (Brownell, 2000; Wing, 1998), initial rate of weight loss is often faster and larger with a mean loss of 20 kg over 12 weeks (Wadden, Stunkard, & Brown ell, 1983). H owever, there is some evidence that prescribing faster initial weight loss may not produce better long term outcomes. For example, Wadden and colleague s (1994) randomly assigned women to either a balanced deficit diet (1200 kcal/day) condition for 52 weeks of a behavioral treatment program or to a very low calorie diet (420 kcal/day) for the initial 16 weeks and a balanced deficit diet thereafter. At si x months, women on the very low calorie diet experienced nearly twice the initial weight loss as those on the balanced deficit diet; however, during the following year of extended care, women who were prescribed the very low calorie diet regained significa ntly more weight. At 18 month follow up, the groups no longer differed with regards to net weight loss. Similarly, Toubro and Astrup (1997) prospectively randomized one group of obese women to eight weeks of VLCD (500 kcal/day) and another group to 17 w eeks of a conventional hypocaloric diet (1200 kcal/day) to achieve similar weight losses at varying rates. Those on the VLCD experienced weight loss at approximately twice the
24 rate of the conventional dieters, but groups did not significantly differ in we ight maintenance measured at one year follow up. These results suggest that utilizing severe caloric restriction methods to attain fast initial weight losses may not lead to successf ul long term weight maintenance and explains why VLCDs are not recommende d for weight loss in most individuals (NHLBI, 1998). Moderate Caloric Restriction Following the NHLBI guidelines (1998), a diet no lower than 1,000 kcal/day is recommended for obese women. Numerous studies of lifestyle interventions prescribing moderate c aloric restriction (i.e. 1,000 to 1,500 kcal/day) have demonstrated that participants who lose at a faster rate in the short term achieve greater weight reduction in the long term (Carels et al., 2003; Jeffery et al., 1998; Nackers et al., 2010 ). Specific ally, Nackers et al. ( 2010 ) conducted a secondary data analysis of the TOURS study in which 230 obese women who were prescribed a 1,200 kcal/day diet underwent 24 weeks of lifestyle intervention followed by one year of extended care follow up were classifi ed according to rate of weight loss during the first month of treatment. Results indicated that those who lost at the fastest rate achieved both greater short and long term weight reductions and were not at increased risk for weight regain than those who lost at a slower rate. In addition, those who decreased weight at a fast rate were 5.1 times more likely to achieve a 10% weight loss at 18 months than the slow losers. Nackers and colleagues ( 2010 ) theorized that, from a learning perspective, losing at a slow initial rate may be less reinforcing to participants than reducing weight at a fast rate. Reinforcers, such as improvements in quality of life factors (e.g. general appearance, body image, physical mobility, energy, and perceived health) achieved
25 early in behavioral treatment for obesity have been associated with greater long term weight loss outcomes whereas consequences, such as unsatisfactory early weight loss, have been associated with poor treatment outcomes (Carels et al., 2003). This sugges ts that when shaping of healthy behaviors occurs slowly, the resulting small weight changes may not serve as sufficient reinforcers for learning and long term habit change necessary to achieve and maintain long term weight l oss success (Nackers et al., 201 0 ). Similar to VLCDs, participants who are prescribed moderate caloric restriction and lose weight at a fast initial rate may also be susceptible to weight regain. Jeffery et al. (1998) conducted a randomized trial in which 130 men and women participate d in an 18 month weight loss treatment program with an additional 12 month follow up. Participants were then categorized into tertiles according to the maximum amount of weight loss. Those participants in the highest maximum weight loss category lost wei ght at more than twice the initial rate of those in the lowest tertile. The results indicated that 23% of those participants who achieved the highest rate of weight loss (i.e. highest tertile of maximum weight loss [ mean loss of 0.68 kg/week]) attained a clinically significant 10% reduction in body weight at 30 months compared to only 9% of those who lost a t a moderate rate (i.e. middle tertile [ mean loss of 0.58 kg/week]) and 2% who lost the slowest (i.e. lowest tertile [ mean loss of 0.29 kg/week]). Howe ver, the results also demonstrated that those who had the largest rate of initial loss experienced larger and more rapid weight regain than those who initially had slower, smaller losses. Similar findings have documented that a large initial weight loss i s a risk factor for
26 weight regain (McGuire et al., 1999; Sbrocco et al., 1999; Wadden et al., 1994; Weiss et al., 2007; Wing & Hill, 2001). Mild Caloric Restriction kcal/day, and thu s suggest that small daily behavioral changes that decrease energy intake by this amount would be sufficient to prevent gradual weight gain and promote sustainable weight loss. an environment that promotes increases in energy intake and decreases in physical activity, small behavioral changes are more feasible to achieve and sustain than larger ones. T heorists argue that not only are sl ight increases in physical activity (i.e. walking 2 000 ex tra steps to burn approximately 100 kcal) and small substitutions in food intake (i.e. diet soda instead of regular soda to save approximately 150 kcal per 12 oz serving) more sustainable, but these minor modifications can increase self efficacy, promoting additional small changes that could eventually lead to larger maintainable changes (Hill, 2009; Hill et al., 2003) Resea rchers have suggested that the small changes approach may play an important role in weight regulation. Because it is often difficult to maintain the degree of caloric restriction promoted by traditional behavioral treatment programs for a prolonged period of time (Schlundt, Sbrocco, & Bell, 1989), Sbrocco and colleagues (1999) decision making model (Sbrocco & Schlundt, 1998), behavioral choice treatment teaches participants to use mild caloric restriction, reducing caloric intake from an estimated 2 500 kcal/day to 1 800 kcal/day, and to make healthy decisions regarding
27 eating choices without considering their eating patterns as dieting. Weight loss is therefore expected to be slower initially, but continuous as participants maintain new decision making regarding healthy eating patterns. Sbrocco et al. (1999) conducted a 13 week randomized trial where obese women took part in either a traditional behavioral treatment that encouraged significant caloric reduction so as to achieve a substantial initial weight loss, or a behavioral choice treatment that promoted mild caloric restriction and a slow, yet steady weight loss. Results indicate d that at both mid and post treatment, those in the traditional behavioral treatment achieved significantly greater weight losses; however, at three month follow up, t his difference was no longer apparent. The behavioral choice treatment group continued to lose weight and achieved significantly greater weight losses at both 6 and 12 month follow up ( 7.0 and 10.1 kg, respectively) than the traditional behavioral trea tment group, which experienced continued weight regain from 6 to 12 months ( 4.5 and 4.3 kg, respectively). Similarly, Lutes and colleagues (2008) conducted a weight loss trial in which 59 adults participated in 16 weeks of either a standard treatment, ASPIRE program, or control condition. The two former condition s received 40 to 45 minute aerobic and resistance training twice per week while the control group was encouraged to make no changes in their usual exercise or dietary intake. The standard tre atme nt group received weekly individual didactic education and behavioral counseling from a nutritionist and they were encouraged to consume no more than 1,600 kcal/day for women and 2,200 kcal/day for men and participate in at least 30 minutes/day of phy sical activity on most days of the week. Those in the ASPIRE program met
28 individually on a weekly basis with a lifestyle coach and were encouraged to set nutrition and physical activity goals involving the small changes that were presented as choices each week (i.e. reducing caloric intake by 200 to 600 kcal/day and slowly increasing physical activity to reach 3,000 steps/day greater than baseline or a total of 10,000 steps/day). Results demonstrated that participants in the ASPIRE program compared to th e standard treatment and control condition s attained a greater initial weight loss ( 4.5 + 3.4 kg versus 1.1 + 2.7 kg versus 0.1 + 2.4 kg, respectively) at 16 weeks (Lutes et al., 2008) In addition, the ASPIRE group sustained the loss and achieved super ior weight reduction at 3 month follow up compared to the standard treatment group ( 4.1 + 5.8 kg versus 1.3 + 2.5 kg ). Lutes and colleagues (2008) suggest that small changes used to induce a slow and steady initial weight loss produce d larger long term weight loss that wa s not susceptible to regain However this study did not incorporate a traditional lifestyle intervention program that utilized cognitive behavioral techniques like goal setting, self monitoring, and problem solving, which may explain t he minimal amount of weight loss achieved by the standard treatment group. In addition, the researchers did not measure mean caloric intake at baseline, posttreatment, or 3 month follow up, and therefore it remains unclear whether participants were consum ing the actual caloric amount prescribed. In sum mary many behavioral treatment programs have proven effective in promoting clinically significant weight losses, but long term weight reduction is difficult to maintain. Evidence suggest s that initial wei ght loss may be associated with long term
29 success; however, it remains u nclear whether there exists an optimal initial weight loss for effective long term weight management Caloric Intake Prescription Perhaps one of the issue s in determining an optimal initial weight loss arises from the idea that adhering to a specific calorie goal in order to lose at a specific rate may prove difficult. Two studies have attempted to answer the question of whether prescribing a mild or moderate caloric restriction pro duces greater long term success in weight management (Das et al., 2008; Toplak, et al., 2005). Das and colleagues (2008) assigned 46 participants to a group of either mild 10% energy restriction with the goal of producing a gradual weight loss or moderate 30 % energy restriction to produce a faster rate and greater overall weight loss. Participants were provided an initial 24 week supply of food and were then asked to maintain their dietary regimen for an additional 24 weeks without food provision. Result s indicated that while the majority of participants achieved a 5 to 10% weight loss, there were no differences in mean percentage weight lost over the 12 month period. In addition, while condition s differed significantly in percent energy restriction dur ing the first 24 weeks, groups did not differ in over the 12 months. This indicates that the 10 % energy restricted group consumed less than prescribed while the 30 % energy restricted group ate more than prescribed. While the aforementioned study utiliz ed the doubly labeled water method of tracking energy intake, similar results were found in a study utilizing self reported dietary intake (Toplak et al., 2005). Toplak and colleagues (2005) assessed whether mild or moderate energy deficit in con junction with pharmacotherapy would produce greater long term weight loss. Participants (N=430) who were randomized to either a
30 500 kcal/day o r 1,000 kcal/day energy deficit significantly differed in self reported caloric intake at 6 months; however, the differenc e was only 111 kcal/day. At 12 months, the mildly and moderately restricted condition s were not consuming significantly different energy amounts. Results indicate that the majority of participants achieved a 5 to 10% weight reduction, but that groups did not differ in regards to weight loss. Toplak and colleagues (2005) suggest ed that participants tend to gravitate toward dietary intake they find manageable regardless of energy prescription. Self Efficacy and Quality of Life Low levels of self efficac y may explain why participants find adherence to energy prescription difficult Self efficacy is based upon the idea that personal beliefs regarding accomplishment, mastery, and outcome expectancy determine whether a person begins a behavior, what goal sh e will attempt to attain, and the amount of effort exerted (Bandura, 1977). Therefore, estimation of capabilities to perform a behavior, or perceived self efficacy, serves as a major determinant of performance independent of actual skill (Bandura, 1986). Higher self efficacy has been linked to greater success in weight loss (Richman, Loughnan, Droulers, Steinbeck & Caterson, 2001) as well as behaviors associated with weight loss, such as dietary adherence (Linde, Rothman, Baldwin, & Jeffery, 2006). War ziski and colleagues (2008) examined self efficacy in relation to eating behaviors to determine if changes in self efficacy were associated with weight loss following an 18 month behavioral intervention. Using the Weight Efficacy Lifestyle questionnaire, a 20 item questionnaire measuring self efficacy specific to eating behaviors ( Clark, Abrams, Niaura, Eaton, & Rossi, 1991), the researchers found that self efficacy increased over
31 the course of treatment and was associated with weight loss at 18 months (Wa rziski et al., 2008). Conversely, poor eating self efficacy has been linked to weight cycling (Kensinger, Murtaugh, Reichmann, & Tangney, 1998) I mprovements in quality of life have also been positively correlated with greater long term weight loss outc omes (Carels et al., 2003; Kolotkin, Crosby, Williams, Hartley, & Nicol, 2001) while weight gain has been associated with reductions in quality of life (Engel et al., 2003). Disinhibition, Restraint Hunger, Depression, and Binge Eating Perhaps another reason partic ipants experience difficulty adhering to prescribed calorie goals is the environmental pressure that promotes increased energy intake and decreased physical activity (Hill et al., 2009). Herman and Polivy (1980) characterize eating patterns a s a balancing act between an innate physiological desire for food and efforts to resist that desire within the environment They coined this cognitively mediated effort Studies have indicat ed that lower levels of cognitive restraint are often associated with higher energy intake (Stunkard & Messick, 1985), while higher levels of restraint are related to lower body weights (Foster et al., 1 998) Researchers have argued that restrained eaters between periods of dieting and overindulgence (Herman & Polivy, 1980; Ruderman, 1986). The latter period of overeating has been hypothesized to result from the cognitive, emotio nal, or pharmacological disruptions in dietary self control, in which the physiological drive for food temporarily prevails (Polivy & Herman, 1983). In experimental conditions, this phenomenon has been termed disinhibition because disruptions in eating pa tterns disinhibit self imposed cognitive restraint of eating
32 behavior (Herman & Polivy, 1980) In experimental settings, it has been well established that restrained eaters who are forced to consume high caloric foods subsequently increase laboratory cons umption of food (Herman & Mack, 1975), potentially due to violation of an all or nothing attitude toward dieting in which one slip is indicative of abandoning the diet (Ruderman, 1986). Therefore, it is not surprising that lower levels of cognitive restr aint have been linked to higher levels of disinhibition et al. 2001; Marcus et al 1985; Wadden, Foster, Letizia, & Wilk, 1993). Stunkard and Messick (1985) created the 51 item Eating Inventory, previously known as the Three Factor Eating Quest ionnaire, a well validated measure of re straint (degree of conscious control over eating) disinhibition (susceptibility to loss of control over eating) and hunger (susceptibility to eat in response to hunger) Das and colleagues (2008) found no differen ce in reported hunger between the mild and moderate energy restricted groups but noted that participants in the mild energy restricted group who rep orted greater disinhibition at six months experienced poorer weight outcome at 12 months. This suggests th at a small change approach to dietary restriction may still expose participants to discretionary foods high in caloric content that increase risk of disinhibition and non adherence to the prescribed diet (Das et al., 2008). Hunger may also contribute to n on adherence to caloric diet. In obese participants prescribed a VLCD, those who endorsed the highest level of hunger consume d three times as many c alories as those who endorsed the lowest level of hunger (LaPorte & Stunkard, 1990). In addition to decre ases in caloric intake adherence, restraint, disinhibition, and hunger have been examined in the context of weight ma nagement Increases in dietary
33 restraint and decreases in disinhibition and perceived hunger have been linked with greater weight loss fol lowing treatment (Foster et al., 1998). Wing and colleagues (2008) recently investigated predictors of weight regain in successful weight losers and found that increases in disinhibition and hunger were related to greater weight regain Also, t hose part icipants who did not receive in person extended care follow up experienced decreases in restraint and related weight regain. During follow up, increases in reported depressive symptoms were associated greater weight regain, suggesting negative affect alon g with uncontrolled eating tendencies may result in difficulties with long term weight loss maintenance (Wing et al., 2008). In addition t o reduced long term maintenance, uncontrolled eating tendencies may also increase other adverse outcomes. For ex ample, the experimental term d isinhibition is often considered analogous to the clinical term of binge eating (Wardle & Beinart, 1981) Between 25 and 50% of obese persons have been found to binge eat (Bruce & Wilfley, 1996), and those who do have also be en shown to demonstrate lower levels of self efficacy and restraint during eating than ob ese non binge eaters (Kensinger et al., 1998) et al. 2001) and greater levels of depression (Polivy & Herman, 1976 ; Polivy & Herman, 1985 ) T herefore relationship s have been illustrated between decreases in cognitive restraint, increases in disinhibition hunger, and depressive symptoms reduced dietary adherence, more frequent episodes of binge eating and greater weight regain et al., 2001 ; Das et al., 2005; LaPorte & Stunkard, 1990; Wing et al., 2008 ).
34 Improving Caloric Intake Adherence Numerous approaches have been utilized to regulate uncontrolled eating tendencies in order to improve caloric adher ence and long term weight management outcomes. These include self monitoring, partial meal replacements, portion controlled meals, structured meal plans, and cognitive behavioral strategies. Each approach is described in detail below. Self M onitoring S el f monitoring or the systematic observation and recording of target behaviors (Kanfer, 1970) has been associated with reduced food intake and successful weight control (Baker & Kirschenbaum, 1993). In fact, self monitoring has consistently been proven to be one of the most important components of weight management, with monitoring foods eaten, time of consumption, quantity eaten, and amount of fat grams consumed all correlated with weight loss (Baker & Kirschenbaum, 1993; Boutelle & Kirschenbaum, 1998). As participants in lifestyle interventions improve the quality and consistency of self monitoring, weight loss increases (Baker & Kirschenbaum, 1993; Boutelle & Kirschenbaum, 1998). While self monitoring may serve as one of the key behavioral strateg ies t o observing caloric intake and improv ing weight outcomes the validity of self monitoring records remains questionable. P articipants in lifestyle interventions tend to underreport di etary intake by approximately 27 to 46% when eating a diet of conventiona l foods (Johnson, Friedman, Harvey B erino, Gold, & McKenzie, 2005; Lichtman et al., 1992; McKenzie, Johnson, Harvey Berino, & Gold, 2002) a discrepancy that may be caused by lack of understanding of portion sizes or forgetfulness in recording foods and be verages (Blundell, J.E., 2000). T herefore,
35 additional behavioral strategies that reduce personal estimation of caloric intake and increase adherence to caloric prescriptions are warranted. Partial Meal Replacements Partial m eal replacements, or commercial ly available products fortified with minerals and vitamins sold as replacement s for one or more meal have been widely prescribed by both health professionals and researchers to improve weight loss success (Egger, 2006; Heymsfield, van Mierlo, van der Knaa p, Heo, & Frier, 2003 ; Keogh & Clifton, 2005 ). These liquids, powders, or snack bars are often supplemented with fresh fruits and vegetables or at least one meal of normal food consumed as part of a low energy diet (Keogh & Clifton, 2005). I n the Look AH EAD study, participants were able to accept or decline use of meal replacements. Results indicated that the number of meal replacements consumed in the first six months was significantly associated to weight loss at week 26, with participants in the highe st quartile of meal replacement use experiencing a 4.1 times greater odds of reaching a 10% weight loss than those in the lowest quartile (Wadden et al., 2009). Meal replacements have also been shown to be as effective as medication (LeCheminant, Jacobsen Hall, & Donnelly, 2005) and more effective than dieting alone (Vazquez et al., 2009) at maintaining weight loss over a one year period Meal replacements offer known calorie content, a simplified choice of foods, little to no preparation, and an avoida nce of problem foods (Wadden et al., 2007 ). Participants have also rated meal replacements more favorably in terms of dietary compliance and convenience compared to participants consuming a conventional diet (Noakes, Foster, Keogh, & Clifton, 2004). In a ddition, partial meal replacements have
36 been associated with a lower risk of inadequate dietary intake of several vitamins and minerals compared to traditional food (Ashley et al., 2007) and hunger suppression for up to five hours following consumption (Ro thacker & Watemberg, 2004). While meal replacements are convenient and provide consistent portion control, they also lack benefits of a balanced diet of real foods, such as non traditional nutrients and phytochemicals that are essential for health (Hannum et al., 2004). P articipants may also experience monotony in consuming similar products one or more times per day (Hannum et al., 2004) and therefore additional strategies to improve caloric adherence may be necessary Portion Controlled Meals Another str ategy utilized to increase dietary compliance involves prepackaged mea ls designed for specific ally assigned calorie goals. Jeffery and colleagues (1993) conducted a 20 week lifestyle intervention program on 202 overweight and obese individuals to examine the impact of food provision on weight loss. Participants receiving reduced calorie prepackaged meals for five breakfasts and five dinners throughout the 18 month program in addition to the standard lifestyle intervention program achieved greater weight l osses at 6, 12, and 18 months than those who only participated in the reduced calorie diet lifestyle interve ntion program. Use of frozen portion controlled entrees in a study of 60 women achieved similar results (Hannum et al., 2004). Women who received at least two frozen meals in addition to set servings of foods from the Food Guide Pyramid experienced a 2.0 kg greater weight loss and greater reductions in metabolic risk factors for disease after only 8 weeks of the intervention than those who did not c onsume portion controlled meals (Hannum et al.,
37 2004) Hannum et al. (2004) concluded that prepackaged frozen entrees may prove to be a key factor in weight control because of ease of preparation, known calorie content, and portion control. No additional weight loss benefit has been found when portion controlled meals are received free of charge compared to when participants purchase these meals on their own (Wing et al., 1996). Structured Meal Plans Win g and colleagues (1996) also found that providing structured meal plans and grocery list s improves the outcome in behavioral treatment. G roups in a lifestyle intervention who received structured meal plans with or without food provision demonstrated greater adherence to self monitoring, more regular eati ng patterns and increased ease in estimating portion size, finding time to plan meals, and restraining from eating when not hungry. Researchers concluded that menus with and without food provision lead to greater likelihood that low calorie foods are avai lable in the home, more structured eating patterns, increased knowledge of caloric content in commonly consumed foods, and greater ease of following a prescribed diet (Wing et al., 1996). Cognitive Behavioral Strategies Lifestyle interventions also incl ude various cognitive behavioral strategies to increase adherence and improve weight loss outcomes. Participants are taught to control eating and activity stimuli, change negative thoughts associated with eating habits, identify and plan ahead for situati ons that are high risk for slips, problem solve for high risk situations, and learn to cope with potential slips (Marlatt & Gordon, 1980; NHLBI, 1998) ( Stunkard & Messick, 1985) includes ite ms that inquire about control of caloric intake
38 and concerns of weight regain questions that coincide with skills that are frequently taught in lifestyle intervention programs (Wing Tate, Gorin, Raynor, & Fava, 2006 ). In a study of obese women six mont h of cognitive behavioral weight loss treatment was linked to significant increases in restraint and decreases in disinhibition and hunger (Foster et al., 1998) In addition, larger increases in restraint achieved during treatment were associated with lar ger weight losses suggesting training in cognitive behavioral strategies increases cognitive control over eating and greater adherence to a low calorie diet (Foster et al., 1998). In summary, it remains unclear whether prescribing either a mild caloric restriction to induce a gradual initial weight loss or a moderate caloric reduction to generate a fast er and larger initial weight loss results in greater long term weight reduction for participants in a lifestyle intervention program. While previous stu dies have indicated somewhat poor adherence to prescribed calorie goals, there exist numerous strategies that can be utilized to improve caloric adherence in weight manageme nt programs; however, it is unknown if these strategies will be effective. In addi tion, no previous studies have assessed whether prescribing either mild or moderate caloric restriction will result in altered levels of dietary restraint, disinhibition, depression, or hunger, as well as changes in number of binge eating episodes. Speci fic Aims and Hypotheses Th e present study examined the effects of prospectively prescribing two caloric intake goals on weight loss at six months and weight maintenance at 12 months in women ages 25 75 years who underwent a lifestyle intervention for obesi ty The study first aim ed to determine whether assigning participants to a moderate caloric restriction
39 (i.e. 1,000 kcal/day) was associated with larger weight loss at both 6 and 12 months than assigning a mild caloric restriction (i.e. 1,500 kcal/day). It was hypothesized that those women prescribed a moderate caloric restriction w ould achieve a greater weigh t loss at six months and greater maintenance and overall weight reduction at 12 months than those prescribed a mild caloric restriction The study additionally assessed whether there existed a difference in the percentage of participants achieving a > 5% weight reduction at both 6 and 12 months. It was hypothesized that a significantly greater percentage of participants within the 1,000 kcal/day con dition would achieve a > 5% loss at both time points, compared to the 1,500 kcal/day condition. The second aim of this study examine d whether participants adhere d to their meth odology that utilize d self monitoring, structured meal plans, and suggested meal replacements and portion controlled meals It was hypothesized that participants w ould adhere to their caloric prescription and that caloric intake between the two condition s would be significantly different. Finally, exploratory analyse s w ere conducted to assess the impact of the 1,000 and 1,500 caloric prescriptions on self reported levels of dietary disinhibition, restraint, hunger, depression, self efficacy and quality of life and to determine whether a mediation existed between these variables and frequency of binge eating episodes according to caloric assignment.
40 CHAPTER 2 MATERIALS AND METHOD S Research Methods and Procedures The present study was a randomized trial t hat evaluate d the effects of prescribing two calorie goals (1,000 versus 1,500 kcal/day) on weight loss at six months. Weight maintenance at one year was also assessed. Approval was obtained from t he University of Flo rida Institutional Review Boa rd. Participants This study included obese women between the ages of 25 to 75 years who weigh ed between 91 to 136 kg (i.e. 200 to 300 lbs). Weight limitations we re imposed according to the energy requirements so that women who we re prescribed a 1,500 kcal /day diet should have lost approximately 0.5 kg ( 1 lb ) per week and women eating a 1,000 kcal/day diet should have lost 0.9 kg ( 2 lb ) per week (NHLBI, 1998). Women were excluded if they lost an excess of 4.5 kg (10 lbs) within the preceding six months. I n addition, participants were required to obtain written consent from their primary care physician stating they ha d no major medical complications that would limit their participation in a weight loss intervention or put them at risk for adverse events. W omen also affirm ed they were not pregnant and agree d to use proper contraception during the course of the study to prevent possible pregnancy. Interested participants were excluded for presence of a major psychiatric disorder or excessive alcohol intake, or if they were unavailab le or unwilling to attend weekly group meetings unwilling to self monitor daily food and caloric intake, unwilling to adhere to the prescribed calorie goal, unwilling or unab le to provide informed consent, unable to read English a t a sixth
41 grade level, participating in another randomized research project, or have participated in a previous in a University of Florida weight loss program. Procedure Recruitment. Participants in the study were recruited through the use of posted flye rs and announcement s in the Gainesville Sun newspaper classified section R ecruitment methods also included direct random mailings to female residents of Gainesville Interested women w ere asked to telephone our office to undergo a brief telephone screen ing questionnaire. If women were eligible following the telephone interview, they were asked to attend an in person assessment visit At this visit prospective participants receive d a de tailed explanation of the study. I nterested participants were ask ed to provide their informed consent. Measurements of height and weight, as well as questionnaires on demographic information, medical history, dietary intake and restraint, physical activity, and self efficacy w ere completed. Potential participants were a lso be given a letter template and asked to obtain their physician approval, indicating they had no medical contraindications to participating in a research study that require d decrease d caloric intake and increase d physical activity and stating that t hey we re not pregnant or planning on becoming pregnant in the next 16 months Participants were then scheduled for a second assessment visit where they returned the physician consent letter and dietary intake questionnaire. Participants were again weighe d at this visit to ensure they had not gained or lost > 10 lbs since their first assessment visit. All physical measures and questionnaires w ere then repeated a t assessments at six months of the intervention and at study conclusion after a total of 1 2 m onths.
42 Lifestyle Intervention The lifestyle intervention consist ed of two phases. Phase I include d 24 week ly group intervention sessions which were held at the University of Florida Participants w ere randomized to a group of either 10 or 30 people. 1 Each session last ed 90 minutes and was designed to decrease caloric intake and increase moderate intensity physical activity Dietary goals involve d reducing energy intake to either 1,000 or 1,500 kcal/day so as to induce either a 0.90 kg or 0.45 kg weigh t loss, respectively. Participants w ere taught and encouraged to maintain a balanced diet according to recommendations from the U.S. Department of Agriculture es to Stop Hypertension consisting of 2 2 to 29 % total kcal from fat, 18 to 21 % from pr otein, and the remaining 55 to 57 % from carbohydrates ( US D epartment of H ealth and H uman S ervices 2005 ) Participants w ere also e ncouraged to increase fruit and vegetable consumption to five servings per day and whole grains to three or more servings per day. In order to increase adherence to caloric intake goals, a variety of strategies w ere utilized. Consistent with the conceptual model from the social cognitive theory (Band ura, 1977, 1986), participant s first receive d education on strategies to control 1 The present dissertation was conducted in conjunction with dissertation research by Pamela J. Dubyak, M.S. She assessed whether participants assigned to a large group size (i.e. approximately 30 peop le) achieved similar weight losses to participants assigned to a smaller group size (i.e. approximately 10 people). Therefore, this study was comprised of four groups consuming 1,000 kcal/day (i.e. one group consisting of 30 participants and the remaining three groups consisting of 10 participants), and four groups consuming 1,500 kcal/day (i.e. again with one group of 30 participants and three groups of 10 participants). Analyses were conducted to assess for interactions between group size and caloric re striction levels. No interactions were found, and therefore results for the present study do not address group size.
43 caloric intake, including structured meal plans, examples of pre packaged portion controlled entrees, and suggestions for partial meal re placements (see Appendix A). Second, self efficacy and positive outcome expectancies were increased as participants were taught successful ways to modify eating behaviors through food tastings Third, participants were guided in setting weekly goals, mo nitoring their progress, controlling negative stimuli, and restructuring maladaptive cognitions in order to improve self regulatory skills. In addition to caloric intake, weekly goals include d assignments that involve d purchasing one new pre packaged meal or planning a menu that incorporates more fruits and vegetables Fourth, problem solving was utilized to overcome barriers to initiating and maintaining behavior change. Physical activity goals were also created each week based on recommendations of the American College of Sports Medicine (Donnelly, 2009). Pedometers were supplied as participants were encouraged to add 3 000 or more steps /day above baseline levels or attain 10,000 steps/ day To accomplish these goals, participants were instructed to ma intain detailed daily written records of their dietary intake and physical activity. Phase I also include d cognitive and behavioral skills trainin g for weight loss consisting of goal setting, self monitoring, self reinforcement, stimulus control, cogniti ve restructuring, and increasing social support. Weekly sessions consisted of a private weigh exercise, feedback and encouragement from group leaders and other group member s, and skills training related to the behavioral strategies for weight loss. Additionally, sessions include d stress and depression coping strategies and techniques for eating away from home.
44 Phase II consist ed of an extended care 6 month follow up progr am in which participants were encouraged to maintain the new eating and exercise habits learned during the Phase I intervention. Participants were also encouraged to continue monitoring dietary intake and physical activity through the use of record ing log s. During Phase II, a ll participants were asked to attend in person group sessions once per month. Measures Body w eight. In the current study, body weight was used as the primary outcome measure. Weight was measured to the nearest 0.1 kg using a digi tal Tanita BWB 800S scale. Participants were weighed at weekly group sessions and at assessments at M onth s 0, 6, and 1 2 while wearing light indoor clothing, without shoes, and with empty pockets. Caloric Intake. P articipants were asked to complete food records daily throughout Phase I and at least three times per week during Phase II of the study. Mean caloric intake was calculated by summing all daily values provided by participants from Months 0 6 and Months 7 12 for Phase I and II, respectively Only self monitoring logs that contained calorie values were included in this analysis. The total number of food records maintained serves as a better predictor of successful weight loss than actual content (Debraganza, 2010 ; Streit, Stevens, Stevens, & Rssner, 1991). For the present study, the total number of daily records completed from Months 0 6 and Months 7 12 was used as a measure of self monitoring adherence. A completed record was operationally defined as having recorded at least two desig nated meals, regardless if caloric value was included.
45 Adherence to Caloric Prescription. Ad herence to prescribed caloric intake goals was calculated by dividing the mean weekly amount consumed by the amount prescribed multiplied by 100. For example, if a participant consumed 900 kcal and her daily prescription was 1,000 kcal, she would be considered 10 % below her prescribed goal Adherence was then classified by the following categorization : 15% or more below the prescribed goal, within (+/ ) 15% of th e prescribed caloric intake goal, and 15% or more above the prescribed goal, based upon classifications utilized in the PREFER behavioral treatment weight loss study, which assessed adherence to kilocalorie goals at baseline, 6, 12, and 18 months (Warziski et al., 2008). Physical activity. F requency and duration of levels of physical activity occurring within the previous week were measured utilizing the short form, self administered International Physical Activity Questionnaire (IPAQ) (Craig, et al., 2 003). A study of reliability and validity demonstrated that the short form IPAQ has reasonable measurement properties at least as good as establish e d self report measures of physical activity (Craig et al., 2003). We examined the change in minutes of mod erate intensit y walking rather than calculate a total score a nd convert to a metabolic equivalent of task (MET). Self Efficacy. The Weight Ef ficacy Life s tyle (WEL) questionnaire was used to measure self efficacy specific to eating behaviors (Clark et a l., 1991). This 20 item questionnaire utilizes a 10 point Likert scale assessing confidence in ability to avoid eating. The WEL contains one total score derived from totaling the scores from 20 items, has demonstrated good psychometric properties and ha s proven to be an
46 acceptable measure of self efficacy judgments for eating behaviors among obese individuals (Clark et al., 1991). Depressive Symptoms. Symptoms of depression were measured by the BDI II, a well validated measure of depressive symptomat ology (Beck, Steer, & Brown, 1996). Disinhibition, Hunger, and Self Restraint. The Eating Inventory/Three Factor Eating Questionnaire (EI) measure s three dimensions of eating behavior: cognitive restraint of eating, disinhibition, and hunger (Stunkard & Messick, 1985). The EI is a 51 item questionnaire that has well established reliability and validit y (Gorman & Allison, 1995). Binge Eating Episodes. T he Binge Eating Scale (BES) a well validated measure assessing binge eating severity in obese populat ions on a continuous scale was utilized to classify binge eaters (Gormally, Black, Daston, & Rardin, 1982). The BES has been found to display good test retest reliability (Greeno, Marcus, & Wing 1995) and moderate associations with binge eating severity measured with food records (Timmerman, 1999) Quality of Life. Quality of life was measured using the well validated MOS Short Form 36 Health Survey (SF 36; Ware, Kosinski, & Keller, 1994). The SF 36 assesses eight domains of health related quality of l ife: role limitations due to physical problems, bodily pain, general health perceptions, vitality, social functioning, role limitations due to emotional problems, and mental health. At least three studies (Fontaine & Barofsky, 2001; Jenkinson, Wright, & C oulter, 1994; Ware et al., 1994) have documented excellent psychometric properties for the SF 36 and validity has been well established in populations of obese adults (Fontaine & Barofsky, 2001).
47 Attendance. Session attendance was recorded if the partici pant arrived at the treatment session and was weighed by a staff member. Group Preference. A self report measure designed for the present study assessed preference to either a 1,000 kcal/day goal in which rate of weight loss is expected to be faster, or t o a 1,500 kcal/day goal where rate of weight loss is predicted to be slower (see Appendix B). Statistical Analyses Primary Aim A sample size of at least 108 participants was selected to provide a statistical power of 0.80 to detect a 3.0% difference in we ight regain between condition s using a 5.5% standard deviation within condition s (two tailed testing with Bonferroni adjustments) as has been demonstrated in previous trials ( Jeffery et al., 2000; Wadden, Berkowitz, Sarwer, Prus Wisniewski, & Steinberg, 20 01 ; Wadden et al., 2005 ) A missing not at random (MNAR) approach was used to examine the data. Differences in weight between Months 0, 6, and 12 were analyzed using pattern mixture models (Little, 1994) A 0.3 kg/month was added to the last recorded in termittent weight for each participant who did not co mplete the assessment at 6 or 12 months Based on this rate, 1.8 kg was added to the Month 6 weight for those who did not comple te Phase II extended care or 12 month assessment. Th is scenario is based on the documented pattern of weight regain following lifestyle treatment (Jeffery et al., 20 00; Perri et al., 2008; Wadden et al., 2001; Wadden et al., 2005). An additional Chi Square analysis was also conducted to assess whether treatment condition s dif fered in those achieving a > 5% weight loss at both 6 and 12
48 months of the intervention. Percent weight change for participants at Month 6 was calculated using the following formula: ((Month 6 weight Month 0 weight) / Month 0 weight) 100. Similarly, Month 12 percent weight change was calculated by the following: ((Month 12 weight Month 0 weight) / Month 0 weight) 100. Secondary Aim In order to examine whether participants first differed with respect to caloric intake goals, weekly reported ca loric intake was averaged across Phase I of lifestyle intervention I ndependent samples t test s were conducted to assess whether the two treatment conditions differ ed in levels of caloric intake according to mean values reported on self monitoring logs Adherence to caloric prescription goals was also assessed for participants in both treatment conditions. Using mean reported calorie intake from self monitoring records during Phase I a dherence w as categorized by consumption 15% or more below prescribe d amount, within (+/ ) 15% of prescribed amount, and consumption 15% or more above prescribed amount These values were based upon calculations dividing the amount consumed by the amount prescribed multiplied by 100. A Chi Square analysis using an alpha level of 0.05 was conducted to assess whether the two treatment conditions significantly differed in categorized levels of adherence during Phase I Tertiary and Exploratory Aims Results of the E ating I nventory w ere utilized to calculate levels of disi nhibition, restraint, and hunger. In addition, data from the BDI II and W eight Efficacy Lifestyle questionnaire provide d levels of depression and self efficacy, respectively. T o assess
49 whether differences between treatment conditions exist ed t hree 3 x 2 repeated measures ANOVA w ere conducted to compare changes in ratings of these constructs from M onth s 0 to 6 M onth s 6 to 12, and M onth s 0 to 12 The between condition variable was treatment condition, while the within condition variable was time. Given the large number of statistical tests conducted, an alpha of 0.01 was used as a conservative balance between Type I and Type II error T o assess whether treatment condition s differ ed in frequency of binge eating episodes, a 3 x 2 repeated measures ANOVA was also conducted utilizing results of the B inge Eating Scale as the between condition varia ble and time as the within condition variable. An alpha level of 0.0 1 w as once again used for significance testing. To assess whether self reported disinhibition restraint, hunger, depression, and self efficacy serve as mediators between treatment condition and frequency of binge eating episodes, the Baron and Kenny method (1986) was used. determine d whether eac h of the variables in question wa s significantly associated with both treatment condition and frequency of binge eating episodes. If correlations were present, a multiple regression was conducted to dete rmine whether a mediation existed. An alpha level of 0.05 was used for s ignificance testing. Finally, to examine changes in quality of life factors across time, repeated measures ANOVA testing was employed as described above, with quality of life factors as between condition variables. Statistical analysis for the primary a im was conducted using SAS statistical software (SAS Institute, 200 8 ). Analyses for the secondary tertiary, and exploratory aims were conducted using SPSS statistical software version 18.0 ( SPSS, Inc., 2010 ).
50 CHAPTER 3 RESULTS Participants Of the or iginal 182 participants assessed on location 44 were initially excluded. The remaining 138 were potentially eligible for randomization; however, prior to initiation of lifestyle intervention, 13 of these individuals declined to participate. Therefore, t he sample was composed of 125 obese women ( mean + SD age = 52.0 + 10.8 years; unadjusted weight = 104.9 + 10.6 kg; BMI = 37.9 + 3.9 kg/m 2 ) Of these individuals, 60 were assigned to the 1 500 kcal/day goal and 65 were assigned to a 1 000 kcal/day goal T wo women, one from each treatment condition, were removed from the study during Phase I due to exclusionary medical conditions and therefore were not included in the study analyses. Overall, 112 individuals completed the assessment at six months (90% of t hose who initiated treatment) and 111 finished the 12 month assessment ( 89 % of those who initiated treatment) See Figure 3 1 for a Consolidated Standards of Reporting Trials ( CONSORT ) diagram which documents participation. No significant differences in baseline age, weight, BMI, race/ethnicity, education, or household yearly income were found between the 1 000 or 1 500 calorie conditions (Table 3 1). There existed a difference in baseline level of physical activity, as measured by the IPAQ, with those women in the 1 000 kcal/day condition endorsing a lower level of physical activity than women in the 1 500 kcal/day condition X 2 (2) = 7.05, p = .029 V = .24. Therefore, analyses were run both with and without level of physical activity as a cov ariate There were no differences in results when the covariate was included and so the following will present results without using physical activity as a covariate.
51 Primary Aim Weight Change Outcomes There existed a significant condition by time int eraction effect across the course of the study, F (2, 122) = 8.39 p < .001 Figure 3 1 For those participants assigned the 1,000 kcal/day goal, there was a significant weight change from Month 0 6 ( t (122) = 12 85, p < .001 d = .96 ) and Month 0 12 ( t (122) = 8.27, p < .001 d = .72 ) This condition also experienced a significant weight regain from Month 6 12 of the trial t (122) = 3.31, p = .001 d = .17 Participants prescribed the 1,500 kcal/day goal also experienced a si gnificant change in weight from Month 0 6 ( t (122) = 7.17, p < .001 d = .56 ) and Month 0 12 ( t (122) = 5.33, p < .001 d = .49 ); however, there was no significant time effect between Month 6 12 for this condition p = .465. Adjusted a nd unadjusted w eight change outcomes (means and SDs ) for both the moderate and mil d caloric restriction condition s from Month 0 6, Month 6 12, and Month 0 12 are presented in Table 3 2 It was found that rates of dropout, eight women in the 1,000 k cal/day condition and six women in the 1,500 kcal/day condition, did not significantly affect the primary study outcomes ( p = .36 0 ). Participants attended a mean 16.9 + 6.5 sessions out of a total 24 possible during Phase I of the trial and 3.0 + 2.3 sess ions our of a total 6 possible during Phase II. While higher rates of attendance were correlated with weight change from Month 0 6 ( r = .58, p < .001) and Month 0 12 ( r = .25, p = .009), there existed no significant differences in attendance rates b etween conditions during Phase I or Phase II, p =.377 and p = .670, respectively (See Table 3 3). Attendance rates during Phase s I and II of the study did differ significantly by group leader ( 15.6 + 6.8 versus 18.4 + 5.8; F (1,121 ) = 6.153 p = .014 2 partial = .05 for Phase I and 2.7 + 2.2
52 versus 3.5 + 2.2; F (1, 121) = 4.706, p = .032 2 partial = .04 for Phase II ) ; however, there was no interaction effect when treatment condition was added to the model during either Phase I or Phase II p = .744 an d p = .897, respectively. There was no effect for condition preference in that participants who were matched to their preferred caloric prescription goal did not achieve significantly different amounts of weight loss at 6 and 12 months than those who were matched to the goal they did not prefer p = .128 and p = .072, respectively; see Table 3 4. Again, there was no interaction effect when treatment condition was added to the model p = .155 at Month 6 and p = .111 at Month 12 Weight Change at Month 6 At the conclusion of the initial six month treatment period the 1,000 kcal/day and 1,500 kcal/day condition s lost an estimated mean ( + S D ) 10.8 + 6.8 and 6. 3 + 6.8 kg respectively At Month 6, the magnitude of these weight reductions significantly dif fered from each other t (122) = 2.05 p = .0 4 d = .66 Of those women who completed the assessment at Month 6, 64% (n = 79 ) attained a weight loss > 5%. Within the 1,000 kcal/day condition, 7 2 % (n = 4 6 ) of participants reached this percentage of loss compared to 56 % (n = 33) in the 1,500 kcal/day condition There existed no significant difference between treatment conditions in proportion of participants achieving a > 5% weight reduction at Month 6 p = .065 Weight Change at Month 1 2 From conc lusion of Phase I of behavioral treatm ent to 12 month follow up, amount of weight regain was assessed by condition Those women assigned the 1,000 kcal/day intake regained an estimated mean 2.1 + 5.2 kg body weight, while those
53 women assigned the 1,500 kc al/day goal regained 0.5 + 5.1 kg body weight. Therefore, overall those prescribed the 1,000 kcal/day intake lost a n estimated total of 8.6 + 8.4 kg at Month 12, while the women prescribed the 1,500 kcal/day intake lost 5.8 + 8.4 kg. These weight losse s no longer significantly differed at Month 12, p = .261 However, given our large standard deviation at Month 12 for the 1,000 and 1,500 kcal/day conditions this study may not have been adequately powered to address differences of this effect size. At Month 12, 52 % (n = 64) of the participants achieved a > 5% weight loss. Within the 1,000 kcal/day condition, 61 % (n = 3 9 ) of participants reached this percent reduction compared to 4 2 % (n = 25) in the 1,500 kcal/day condition. S ignificantly more women i n the 1,000 kcal/day condition achieved a weight change > 5% at Month 12 X 2 (1) = 4.239 p V = .19. Secondary Aim Mean self reported caloric intake for Phase I of the lifestyle intervention was calculated by averaging week ly reported cal oric intake during the 24 week period. The participants prescribed a 1,000 kcal/day goal consumed a mean 1,164 + 170 calories, while those participants prescribed 1,500 kcal/day consumed a mean 1 ,518 + 22 2 calories Mean caloric intake was found to be si gnificantly different between conditions t (122) = 10.01, p < .001 d = 1.80 T o assess whether participants in each condition were adherent to their p rescribed caloric intake goals, caloric intake percentages were calculated by dividing partici mean caloric intake by their prescribed goal and multiplying by 100. On average, participants in the 1,000 kcal/day condition consumed 16% more calories than
54 prescribed Women who were prescribed the 1,500 kcal/day intake goal ate 1% more calories than prescribed Using adherence categories where participants we re classified according to percent above or below their prescribed amount (e.g., 15 % or more below prescribed goal, within (+/ ) 15% of prescribed amount and 15% or more above caloric intake goal) results indicated significant differences between the treatment conditions during Phase I X 2 (2) = 17.2, p < .001 V = .37 While 77 % of the women prescribed the 1,500 kcal/day diet were classified as within 15% of their prescribe d amount, 53% of those instructed to consume a 1,000 kcal/day diet were found to be within this range. In addition, 45% of participants in the 1,000 kcal/day condition were categorized as consuming > 15% above their prescribed amount. Ten percent of the women on the 1,500 kcal/day prescription were categorized as consuming at least 15% b elow their prescribed amount. See Table 3 5 for categories based on percent adherence by condition A trend was observed between p articipant adherence to caloric prescri ption and weight change as those who adhered within (+/ ) 15% of their caloric intake goal lost 9.6 + 7.2 kg compared to the 7.0 + 6.8 kg reduction achieved by those who did not consume within 15% of their prescribed calorie goal p = .058. An addition al Independent S amples t test was conduct ed to assess whether conditions differed in the number of self monitoring records kept during Phase I A self monitoring log was considered complete if at least two designated meals were recorded in one day. Out o f a potential 168 records, women in the 1,000 kcal/day condition turned in 98.3 + 49.9 food records, whi le those in the 1,500 kcal/day prescription turned
55 in 98.7 + 57.4 self monitoring logs. The condition s did not differ in respect to adherence to mainta ining self monitoring logs p = .973. During Phase II of the trial, women were asked to maintain at least three days of food records per week. Therefore, the total number of weeks of records was u sed to assess differences in adherence to self monitoring goals between conditions Out of a total potential 24 weeks of self monitoring logs, those prescribed 1,000 kcal/day returned 4.7 + 7.2 weeks of records, while those prescribed 1,500 kcal/day maintained 6.8 + 9.7 weeks of records p = .182 While no betw een condition differences existed this variable was significantly correlated to weight change from Months 6 to 12, with those participants maintaining a greater number of food records achieving greater weight loss success r = .44, p < .001 During Pha se II participants were asked to maintain their prescribed caloric intake goals un less they had achieved a BMI of < 25, which pertained to one participant in the 1,000 kcal/day condition During the extended care period, partici pants within the treatment conditions continued to consume significantly different levels of caloric intake, with those prescribed 1,000 kcal/day reporting 1,247 + 246 kcal/day and those prescribed 1,500 kcal/day reporting 1,488 + 208 kcal/day t (56) = 4.00, p < .001 d = 1.06 Tertiary and Exploratory Aims R estraint, Disinhibition, Hunger Table 3 6 outlines values of restraint, disinhibition, and hunger specific for the 1,000 kcal/day and 1,500 kcal/day conditions at Months 0, 6, and 12. Restra int was found to incre ase from M onth 0 to M onth 6 and sligh tly decrease at Month 12 p < .01) and therefore
56 Greenhouse Geisser results were utilized. There existed a significant main effect for restraint across the three time poin ts F (1.78 169.31 ) = 151.51, p < .001 2 partial = .70 However, the interaction effect of caloric prescription across time was not significant p = .978, indicating the conditions did not differ in reported levels of restraint throughout the study period. Levels of disinhibition decrease d wit hin the sample from baseli ne to end of Phase I and then slightly increase at the end of Phase 2 There existed a significant main effect for changes in level of disinhibition across time, F (2 186 ), = 52.87, p < .001 2 partial = .55 ; however, l evels of d isinhibition did not differ between conditions across the three time periods p = .188. H unger also decreased from Month 0 to M onth 6 and remained f airly stable at Month 12 p < .01) and theref ore Gre enhouse Geisser values wer e used. The changes in hunger level was found to be significant across time, as indicated by a significant main effect F (1.82 169.38 ) = 56.28, p < .001 2 partial = .55 T here was not a significant interaction effect of caloric prescription, indicating the conditions did not differ from each other across the study period with respect to levels of hunger p = .346 Depression Levels of depression at b oth 6 and 12 months did not meet the assumptions for normality with regards to skewness (z = 2.57 at M onth 6 and z = 2.34 at M onth 12). Because the depression variables were positively skewed, they were transformed using a square root i + 1) for subse quent analyses. Participants exhibited minimal levels of depressive sympto m s at the three assessment points. Trends from the overall
57 sample demonstrated a significant decrease in depressive symptoms during Phase I and a slight increase by M onth 12. A si gnificant main effect for d epressive symptoms existed across time, F (2 188 ) = 37.37, p < .001 2 partial = .45 ; ho wever, the interaction effect of time by condition was not significant, indicating conditions did not differ in levels of depression across Month 0, 6, or 12 of the study p = .607; see Table 3 6 Self Efficacy L evels of self efficacy we re found to increase from Month 0 to 6 and slightly decrease at Month 12 Table 3 6 outlines values specific for the 1,000 kcal /day and 1,500 kcal/day conditions efficacy, and thus the Greenhouse Gei sser values were utilized for the results. A significant main effect indicates reported levels of self efficacy differed across the three assessment points F (1.72 166.33 ) = 55.08, p < .001 2 partial = .49 Again, conditions did not differ significantly across the course of the study, as indicated by a non significant interaction effect between time and treatment condition p = .482 Mediation of Binge Eating It was found at the three ass essment points, the mean reported level of binge eating did not classify our sample as binge eaters (as demonstrated by a score > 17 on the BES). Binge eating decrease d from M onth 0 to M onth 6 and slightly increase d at M onth 12 This trend was consistent for both the 1,000 and 1,500 calorie conditions (see Table 3 6 ). Results demonstrated a significant main effect for time with respect to levels of binge eating F (2 190 ) = 51.32, p < .001 2 partial = .49 ; however, the interaction of condition s a cross time was not significant p = .072. Because condition s did not differ on levels of disinhibition, restraint, hunger, depression, self efficacy, or with respect to
58 frequency of binge eating epis odes, a mediation analysis of associations between these variables according to caloric assignment could not be conducted. Quality of Life Quality of life was measured using the nine subscales of the SF 36 : Reported Health Transition, Physical Function ing, Role Functioning Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Functioning Emotional, and Mental Health Table 3 7 provides values for the 1,000 kcal/day and 1,500 kcal/day condition s as well as the total sample on each of these subscales Repeated measures 3 x 2 ANOVAs were conducted to assess for changes in reported levels of quality of life between treatment conditions No significant main effect for time or significant interaction effect s with treatment condition were observed for levels of Role Functioning Physical, Bodily Pain, Social Functioning, or Role Functioning Emotional. Functioning, General Health, Social Functioning, and Mental He alth, and therefore the Greenhouse Geisser results were utilized. There existed significant main effects across time for Health Transition F (1.69 153.86 ) = 43.17, p < .001 2 partial = .40 Physical Functioning F (1.78 170.70 ) = 17.57, p < .001 2 part ial = .28 General Health F (1.83 175.84 ) = 16.29, p < .001 2 partial = .26 Vitality F (2 192 ) = 26.96, p < .001 2 partial = .36 and Mental Health F (1.80 172.48 ) = 9.82, p < .001 2 partial = .20 There existed no significa nt interaction effects b etween caloric prescription and time on any quality of life variables except Social Functioning Results of an Independent Samples t test indicated condition s significantly differed on reported levels of Social Functioning at Month 12, t (97) = 2.67, p = 009 d = .54
59 Weight Regain It was found that 28 participants regained > 4.5 kg body weight from 6 to 12 months, thus contributing to the large standard deviation at Month 12. Sixteen of these women w ere p rescribed a 1,000 kcal/day goal, while 1 2 were assign ed the 1,500 kcal/day condition Exploratory Analysis of Variance tests were conducted using Month 0 study measures to investigate whether baseline characteristics were associated with weight regain (Table 3 8) Participants who regained > 4.5 kg from Months 6 to 12 reported higher levels of depressive symptoms ( F (1, 116) = 4.439, p = .037 2 partial = .03 ), disinhibition ( F (1, 117) = 5.159, p = .025 2 partial = .04 ), and greater caloric intake during the baseline week of the intervention ( F (1, 121) = 4.232, p = .042 2 partial = .03 ) compared to those who regained < 4.5 kg. When treatment condition was added to the model no interaction effects by caloric prescription were found for depression ( p = .241 ) disinhibition ( p = .357) or baseline caloric intake ( p = .348) No differences existed between regain categories with respect to baselin e age, weight, restraint, hunger, binge eating, self efficacy, or physical activity. In additional exploratory analyse s, weight regain was kept as a continuous variable and baseline levels of depressive symptoms, disinhibition, and caloric intake were cat e gorized. Levels of depression were dichotomized as above or below the cutoff for mild depression (i.e., a score > 14 on the BDI II). It was found that 78 participants endorsed scores below the cutoff (n = 41 in the 1,000 kcal/day condition and n = 3 7 in the 1,500 kcal/day condition) and 40 participants reported mild depression or above (n = 21 in the 1,000 kcal/day condition and n = 19 in the 1,500 kcal/day condition) Given no clinical level exists for determining disinhibition (Stunkard &
60 Messick, 1985 ) this variable was categorized as either above or below the median level the median score was found to be 10 Finally, baseline caloric intake was also dichotomized according to the median of 1882.3 kcal/day as no level of mean baseline energy intake for a population similar to the present study was found These potential moderators were again assessed by treatment condition and weight regain from Months 6 12 ( Table 3 9). In the analysis including depressive symptoms, no significant main effects for treatment condition or depressive symptoms were found p = .427 and p = 167 respectively. In addition, no interaction effect existed between the two variables an d weight change during Phase II p = .698 In the analysis of disinhibition x treatment condition x weight regain, no main effects existed for disinhibition ( p = .436) or treatment condition ( p = .053). Additionally, there did not exist a significant i nteraction effect among these variables p = .169. While no significant main effects existed for baseline caloric intake x treatment condition x weight regain from Months 6 12 ( p = .144 for treatment condition and p = .714 for baseline caloric intake) a significant interaction effect was found among treatment condition x baseline caloric intake x w eight regain during Phase II, F (1, 119) = 4.169, p = .043 2 partial = .03 Post hoc testing indicated that participants with higher baseline caloric intake who were assigned the 1,000 kcal/day condition regained significantly greater amounts of weight than those with higher baseline caloric intake who were presc ribed a 1,500 kcal/day diet, t (60) = 2.488, p = .01 6 d = .64 Exploratory analyses were also conducted to assess for differences during the intervention between those who regained > 4.5 kg and < 4.5 kg (Table 3 10 ) N o
61 significant differences wer e found with respect to comp letion of self monitoring logs ( p = .192) or percent adherence to caloric prescription during Phase I of the intervention ( p = .547) There was, however, a significant difference in number of sessions attended during Phase I. ( F 1, 121) = 8.408, p = .004) indicating a violation of homogeneity of variances, and therefore, Brown Forsythe results are shown. Those who regained large amounts of weight attended more sessions compared to those who did n ot regain a substantial amount of weight F (1, 58.857) = 5.162, p = .027 2 partial = .03 No interaction effect by caloric prescription condition was found p = .254 N o significant differences were found between regain groups in attendance rates during Phase II p = .296 With respect to number of calorie records completed during Phase II F (1, 121) = 20.928 p < .001), therefore the Brown Forsythe test was utilized. Those participants who regained > 4.5 kg returned significantly fewer weeks of calorie records compared to the group who d id not regain 4.5 kg or more ( F ( 1, 74.765 ) = 11.717 p = .001 2 partial = .05 ) suggesting lower adherence to self monitoring in the large regain group. Again, no interaction effect by treatment condition was found p = .567 At Month 6, regain groups did not differ on r eported levels of depression ( p = .924) restr aint ( p = .129) disinhibition ( p = .987) hunger ( p = .985) binge eating ( p = .906) and self efficacy ( p = .291) Reported depressive symptoms at Month 12 were signif icant ( F (1, 99) = 16.028, p < .001) so the Brown Forsythe test was used. Those who regained > 4.5 kg reported greater depressive symptoms than those who regained less F (1, 31.862) = 5.240, p = .029 2 partial = .08 In addition, a ll three variables
62 assessed by the Eating Inventory were also found to be different between regain groups, with those regaining more weight reporting significantly lower levels of restraint ( F ( 1, 99 ) = 8.43 p = .005 2 part ial = .08 ) and higher levels of disinhibition ( F (1, 99) = 7.674 p = .00 7 2 partial = .07 ) and hunger ( F (1, 99 ) = 8.425 p = .005 2 partial = .08 ). Reported levels of self efficacy were found to be significantly lower in those regaining > 4.5 kg ( F (1, 9 9 ) = 12.903 p = .001 2 partial = .12 ) while endorsement of binge eating was significantly higher in the larger regain group compared to those who regained < 4.5 kg, F (1, 99 ) = 12.574 p = .001 2 partial = .11 (See Table 3 1 1 for study measures at Mon ths 6 and 12 according to weight regain category). Again, no interaction was significant for any aforementioned variables when caloric prescription treatment condition was added to assess for a moderation p = .931 for depression; p = .373 for restraint; p = .599 for disinhibition; p = .889 for hunger; p = .858 for self efficacy; and p = .382 for binge eating
63 Table 3 1. Baseline v alues a ccording to c aloric p rescription 1,000 kcal/day 1,500 kcal/day Total sample n = 65 n = 60 N = 125 M SD M SD M SD Age (years) 51.5 11.7 52.3 9.8 52.0 10.8 Unadjusted w eight (kg) 105.0 10.6 104.7 10.7 104.9 10.6 BMI (kg/m 2 ) 38.1 4.0 37.6 3.8 37.9 3.9 n % n % n % Race/ethnicity Caucasian 48 73.8 44 73.3 92 73.6 African American 9 13.8 12 20 .0 21 16.8 Hispanic American 4 6.2 0 0 4 3.2 Other 3 4.6 1 1.7 4 3.2 Declined to respond 1 1.5 3 5.0 4 3.2 Education 12 years or less 3 4.6 3 5.0 6 4.8 13 15 years 33 50.8 28 46.7 61 48.8 16 years or more 29 44.6 29 48.3 5 8 46.4 Household yearly income < $35,000 12 18.5 15 25.0 27 21.6 $35,000 to $49,999 10 15.4 12 20.0 22 17.6 $50,000 to $74,999 21 32.3 12 20.0 33 26.4 > $75,000 18 27.7 18 30.0 36 28.8 Not reported 4 6.2 3 5.0 7 5.6 Ta ble 3 2. Weight c hanges from Month 0 to Months 6 and 12 a ccording to caloric p rescription 1,000 kcal/day 1,500 kcal/day Unadjusted Adjusted Unadjusted Adjusted M S D M S D M S D M S D Month 0 6 11.6 7.3 10. 8 a, 6.8 6. 7 6.1 6. 3 a, 6.8 Month 6 12 2.5 4.9 2.1 b 5.2 0.5 4.7 0. 5 5.1 Month 0 12 9. 5 7.8 8.6 c 8.4 6. 3 8.7 5. 8 c 8.4 Note: a p < .001 for Month 0 6; b p < .001 for Month 6 12; c p < .001 for Month 0 12; p < .05 between treatment conditions for Month 0 6
64 Table 3 3. Attendance rates for Phase I and Phase II a ccording to caloric p rescription 1,000 kcal/day 1,500 kcal/day M SD M SD Phase I (out of 24 total) 19.1 4.5 17.5 6.3 Phase II (out of 6 total) 3.7 1.8 3.3 2.3 Table 3 4. Weight changes a ccording to m atched and u nm atched caloric p rescription p reference and a ssignment 1,000 kcal/day 1,500 kcal/day Total Sample Matched Unmatched Matched Unmatched Matched Unmatched M SD M SD M SD M SD M SD M SD Month 0 6 9.2 6.5 12.1 7.8 6.5 6.4 5.8 6.0 7.7 6. 5 9.7 7.8 Mon th 0 12 6.4 6.7 10.9 7.9 5.8 9.1 5.4 8.2 6.1 8.1 8.8 8.3 Ta ble 3 5 Percent a dherence c ategorization for Phase I a ccording to c aloric p rescription 1,000 kcal/day 1,500 kcal/day n % n % > 15% Below Caloric Prescription 1 1.6 6 10.0 Within (+/ ) 15% of Caloric Prescription 34 53.1 46 76.7 > 15% Above Caloric Prescription 29 45.3 8 13.3 Note: p < .001 between conditions
65 Table 3 6 Restraint, d isinhibition, h unger, s elf e fficacy, d epression, and b inge e ating at Months 0, 6, and 12 a ccording t o c aloric p rescription 1,000 kcal/day 1,500 kcal/day Total Sample M SD M SD M SD Restraint Month 0 7. 3 3.7 8.3 3.9 7.7 3.8 Month 6 13.9 3.3 14.6 3.2 14.2* 3.2 Month 12 12.3 4.0 13.1 3.5 12.7 3.8 Disinhibition Month 0 9.7 3.3 10.1 3. 2 9.9 3.2 Month 6 6. 4 3.0 6. 8 3. 8 6.6 3.4 Month 12 8.3 3. 8 7. 8 4.0 8.1 3.9 Hunger Month 0 7. 3 3. 6 7. 3 3.1 7. 3 3.3 Month 6 3.9 2.9 4. 3 3.3 4.1 3.1 Month 12 5.0 3. 5 4. 6 2. 8 4.8 3.1 Self Efficac y Month 0 108.8 32.2 107. 3 26.2 108. 1 29.4 Month 6 138.9 21.8 137.9 19. 8 138.4 20.7 Month 12 126.9 31.1 12 7.8 26.1 127.4 28.6 Depression Month 0 11.6 10.3 11.0 6.3 11.3 8.6 Month 6 6.0 7.6 5.2 5.8 5.6 6.7 Month 12 5.7 7.2 6.5 7.6 6.1 7.4 Binge Eating Month 0 16.4 8. 3 16. 1 6.8 16.2 7.6 Month 6 7. 9 6.1 8. 9 7.5 8.4 6.8 Month 12 11.3 7. 9 9.4 7. 3 10.4 7.6 Note: p < .001 compared to Month 0 value
66 Table 3 7 Quality of l ife v ar i ables from the SF 36 at Months 0, 6, and 12 a ccording to c aloric p rescription 1,000 kcal/day 1,500 kcal/day Total Sample M SD M SD M SD Health Transition Month 0 3. 1 1.0 3. 1 1. 2 3. 1 1. 1 Month 6 1. 9 1.0 2.0 1.0 1.9 a 1.0 Month 12 2 1 0.9 1.9 1. 1 2.0 b 1.0 Physical Functioning Month 0 69. 9 20.6 72. 7 21. 9 71.2 21. 0 Month 6 83.5 18.0 79.4 23. 4 81.5 a 20.8 Month 12 80.2 23. 7 79.6 24. 3 79.9 b 23. 9 Role Functioning (Physical) Month 0 73.5 34. 8 71. 3 35.6 72.4 35.0 Month 6 83. 2 31. 6 75.9 37. 3 79. 6 34. 6 Month 12 85.5 29.9 75.0 38.8 80. 3 34. 9 Bodily Pain Month 0 68.8 25.3 62.5 23.5 65.5 24.5 Month 6 71.1 22. 4 66.5 23. 9 68.8 23.1 Month 12 67.9 22.0 6 7.0 24. 3 67.4 23. 1 General Healt h Month 0 64.3 22.3 66.2 17.9 65.2 20. 3 Month 6 72.7 19. 9 75.7 15. 6 74.2 a 17. 9 Month 12 69.9 19.5 73.3 18.6 71.6 b 19.0 Vitality Month 0 43.9 23.8 45.7 18. 5 44. 8 21.3 Month 6 61.7 20. 5 60.8 15.1 61.3 a 1 8.0 Month 12 56.4 21. 5 57. 7 19. 6 57.0 b 20. 5 Social Functioning Month 0 72. 9 25.8 73.9 24. 4 73. 4 25. 1 Month 6 84. 8 18. 6 8 2.0 19.7 83.4 19.1 Month 12 88.5 19.7 76.0 26. 4 82.2 24.0 Role Functioning (Emotional) Month 0 72.3 39. 8 65. 6 38 .8 69. 1 39.3 Month 6 73.9 40.4 75.3 35. 6 74.6 37.9 Month 12 85.3 32.4 76. 7 36.4 81.0 34. 6 Mental Health Month 0 71.8 19.4 71.7 17. 2 71. 8 18. 3 Month 6 79. 3 15.9 80. 7 12. 5 80.0 a 14. 3 Month 12 78. 5 17. 6 77. 3 18.3 77.9 b 1 7 9 No te: a p < .0 01 for Month 0 Month 6; b p < .001 for Month 0 Month 12; p < .01 between treatment conditions at Month 12
67 Table 3 8 Month 0 s tudy v ariables a ccording to w eig ht r egain from Months 6 to 12 Regain > 4.5 kg (n = 28) Regain < 4.5 kg (n = 95) M SD M SD Age 51.3 11.5 52.2 10.7 Weight 105.1 11.0 104.7 10.5 Depression 14.3 10.7 10.3 7.7 Restraint 7.4 3.9 7.8 3.9 Disinhibition 11.1 2.4 9.5 3.3 Hunger 7.6 2.7 7.1 3.5 Binge Eating 17.8 6.0 1 5 7 7.8 Self Efficacy 98.8 29.6 111 .1 28.9 Week 1 calorie mean 2016.0 523.0 1712.7 725.8 Note: p < .0 5 between categories Table 3 9. Weight r egain by Month 0 s tudy v ariables a ccording to t reatment c ondition 1,000 kcal/day 1,500 kcal/day M SD M SD Depression Score > 14 1.8 7.2 1.4 4.1 Score < 14 1.8 4.3 0.6 4.4 Disinhibition Score > 10 1.6 6.7 1.1 4.2 Score < 10 2.2 3.6 1.0 5.2 Week 1 calorie mean Level > 1882.3 3.1* 4.8 0.0 5.1 Level < 1882.3 0.9 5.7 1.5 3.5 Note: p < .05 for interaction of treatment condition, week 1 calorie mean, and weight regain
68 Table 3 10 Intervention v ariables a ccording to w eight r egain from Months 6 to 12 Regain > 4.5 kg Regain < 4.5 kg M SD M SD Ph I Attendance (# of weeks) 19.0* 5.0 16.4* 6.7 Ph I Weeks of calorie records 16.2 6.2 14.3 8.1 Ph I Caloric Adherence (%) 110.7 14.4 108.4 18.2 Ph II Attendance (# of weeks) 2.8 1.8 3.8 2.1 Ph II Weeks of calorie records 2.2** 5.4 6.9 ** 9.0 Note: p < .01 and ** p < .001 betw een categories Table 3 1 1 Month 6 and 12 s tudy v ariables a ccording to w eight r egain from Months 6 to 12 Regain > 4.5 kg Regain < 4.5 kg M SD M SD Month 6 Depression 5.5 4.6 5.6 7.4 Restraint 15.0 2.6 14.0 3.4 Disinhibition 6.6 3.0 6.6 3.5 Hunger 4.1 2.8 4.1 3.2 Binge Eating 8.5 5.3 8.4 7.3 Self Efficacy 134.8 22.0 139.6 20.3 Month 12 Depression 9.5 9.7 4.9* 6.0 Restraint 10.9* 3.3 13.3* 3.7 Disinhibition 9.8* 3.3 7.4* 3.9 Hunger 6.3* 3.3 4. 3* 2.9 Binge Eating 14.7* 8.3 8.8* 6.8 Self Efficacy 110.9* 28.0 132.9* 26.9 Note: p < .01 between categories
69 Figure 3 1. CONSORT diagram of participation rates during the 12 month program 182 Assessed Onsite 44 Excluded 28 No showed or cancelled pr ior to second assessment visit 7 BMI > 45 5 Could not attend meetings 4 No physician consent 125 Randomized and initiated lifestyle modification program 13 Declined to participate 10 Could not attend meetings/lost interest 1 Lost to follow up 1 Became pregnant 1 Moved away 138 Potentially eligible for randomi zation 65 Assigned 1,000 calorie goal 60 Assigned 1,500 calorie goal 56 Assessed at M onth 6 4 Lost to follow up 2 Declined/moved away 2 Dropped out 1 Had cancer 56 A ssessed at M onth 6 3 Lost to follow up 1 Had cancer 57 Assessed at M onth12 3 Lost to follow up 2 Declined 2 Dropped out 1 Had cancer 54 Assessed at M onth 12 3 Declined 2 Lost to follow up 1 Had cancer
70 Figure 3 2. Adjusted w eight changes according to treatment condition (Means and Standard Errors) p = .04 b etween treatment conditions at Month 6; a p < .001 for Month 0 Month 6 for both condition s; b p < .001 for Month 0 Month 12 for both condition s; c p = .001 for Month 6 Month 12 for the 1,000 kcal/day condition 90 92 94 96 98 100 102 104 106 108 Baseline Month 6 Month 12 1,000 kcal/day 1,500 kcal/day 104.8 94.1 104.6 98.4 98.8 96.2 kg a a b b, c
71 CHAPTER 4 DISCUSSION Primary Aim The present lifestyle intervention trial examined whether prospectively prescribing a moderate caloric restricti on o f 1,000 kcal/day was associated with larger weight loss at six months and greater weight maintenance at 12 months than assigning a mild caloric restriction of 1,500 kcal/day With regards to the primary aim, there were t hree key findings. First, whil e both condition s experienced a significant change in weight from baseline to six months, women who were assigned to the 1,000 kcal/day goal achieved a significantly greater weight loss than those women who were asked to consume 1,500 kcal/day ( 10.8 + 6.8 versus 6.3 + 6.8, respectively) Wh en categorized according to those who achieved a > 5% weight loss at Month 6, 64% of the sample achieved this level of weight reduction ; however, no differences existed between treatment conditions in the percentage of participants reaching a > 5% loss Second, it has been well documented that weight loss usually slows following six months of behavioral treatment (Jeffery et al., 2000; Wing, 2002), and weight regain begins (Perri, 1998; Wadden et al., 2007). Partic ipants in the present study who were assigned a moderate caloric restriction of 1,000 kcal/day experienced a significant amount of weight regain between Months 6 12, while those assigned a mild caloric restriction of 1,500 kcal/day regained a nonsignific ant amount of weight from end of treatment at six months to 12 month follow up (2.1 + 5.2 kg v ersus 0.5 + 5.1 kg, respectively). The third key finding illustrates that both conditions experienced a significant change in weight from baseline to 12 months. However, as participants in the 1,000
72 kcal/day condition regained a significant amount of weight during the six month extended care period, at 12 months, weight changes were not significantly different between the 1,000 and 1,500 kcal/day conditions ( 8. 6 + 8.4 versus 5.8 + 8.4 kg, respectively) Of note, while participants in the 1,500 kcal/day condition did not regain a large amount of weight, they also did not continue to lose during Phase II of the study as is illustrated in other studies which util ized a small change approach (Lutes et al., 2008; Sbrocco et al., 1999). This condition also remained 2.8 kg heavier than those prescribed the 1,000 kcal/day goal, which stands as a large clinical difference Our initial power analysis indicated a sampl e size of at least 108 participants to provide a statistical power of 0.80 to detect a 3.0% difference in weight regain between conditions using a 5.5% standard deviation within conditions. Given our large standard deviation at Month 12 for the 1,000 and 1,500 kcal/day condition s (8.4 and 8.4, respectively) this study may not have been adequately powered to address differences of this effect size. There did however, exist a significant difference in the number of participants who achieved a weight loss > 5% at Month 12. While 42% (n = 25) participants in the 1,500 kcal/day condition reached this level of weight change, 61% (n = 39) in the 1,000 kcal/day condition achieved this loss. From a clinical standpoint, maintaining a weight change > 5% has been associated with reductions in glycemic control, hypertension, and hyperlipidemia (Diabetes Prevention Program Research Group, 2002; The Look AHEAD Research Group, 2010). Conversely, when weight is regained and net reduction is < 5%, participants often los e the beneficial health effects associated with diabetes risk (Krebs et al., 2002), blood pressure (Stevens et al., 2001), and lipid control (Wadden, Anderson, and Foster, 1999). More participants in the 1,000
73 kcal/day prescription achieved and maintained a > 5% weight reduction, which suggest potential clinical benefit of prescribing a greater caloric reduction and promoting greater initial loss in weight management programs. Based upon classical and operant learning theory, greater initial changes in die t would result in larger reductions in body weight, which then reinforce habit change (Ferster et al., 1962; Stuart, 1967; Wadden et al., 2007; Wing, 2002). This appears congruent with 1,000 kcal/day condition where greater initial change in diet resulted in larger weight loss at six months compared to the 1,500 kcal/day condition These results are similar to those found by Nackers and colleagues (2010) and Carels and colleagues (2003) where greater initial behavioral changes and a posited reinforcing in itial weight loss wa s associated with improved weight loss outcome following a six month lifestyle intervention. However, reducing calories so as to lose larger amounts of weight initially may also increase susceptibility to weight regain (McGuire et al. 1999; Sbrocco et al., 1999; Wadden et al., 1994; Weiss et al., 2007; Wing & Hill, 2001) This appeared especially true for those participants who consumed greater amounts of calories at baseline and who were then placed in the large caloric restriction condition of 1,000 kcal/day. These participants were asked to greatly rest rain their dieting habits According to the disinhibition hypothesis laid out by Herman and Polivy (1980), the level of dietary restraint required to sustain this reduction in cal oric intake may have been too l arge to maint ain, thus resulting in dietary violation While the caloric prescriptions utilized in the current study did not fall under the classification of VLCDs, the weight regain trend seen from 6 12 months within the 1, 000 kcal/day condition followed a similar pattern.
74 Wadden and colleagues (1994) utilized a VLCD of 420 kcal/day to initiate a large initial weight loss during the first 16 weeks of a 52 week behavioral treatment program. While these participants experien ced ne arly twice the initial weight reduction as those assigned a balanced deficit diet of 1,200 kcal/day at six months, they also regained significantly more weight (Wadden e t al., 1994). Typically, VLCDs utilize meal replacements to achieve the prescrib ed energy content of 400 800 kcal/day but participants are not asked to sustain this caloric intake level long term (Wadden et al., 1994; Wadden & Stunkard, 1986; Wing, Blair, Marcus, Epstein, & Harvey, 1994; Wing et al., 1991). The current study differs in that participants were not prescribed a VLCD; however, reported caloric intake for participants in the 1,000 kcal/day condition did increase following the first six months, while the majority of participants within the 1,500 kcal/day group remained adh erent to their caloric prescription. In a review of participant weight loss during lifestyle interventions, Heymsfield and colle agues (2007) concluded that minimal weight change s ass ociated with low calorie diets are not likely due to biological adapta tions, but more so attributed t o difficulties with adherence to caloric intake Kirschenbaum and Tomarken (1982) and Polivy and colleagues (1986) assessed restrained eaters and found that overeating reflected a disruption in self monitoring of eating patt erns Applying this idea to the present study suggests that participants who highly restrained their eating (i.e., those assigned the 1,000 kcal/day goal) may have found it difficult to continuously monitor their eating behaviors in a strict manner. Hill and colleagues (2003) promote small daily reductions in energy intake that are thought to be more feasible to achieve and sustain than larger ones. This concept also supports the notion that, in the present study, those women
75 prescribed the 1,500 kcal/da y goal may have been more likely to attain and maintain their energy prescription, while the women prescribed the 1,000 kcal/day may have initially achieved their energy intake goal, but were unable to sustain the behavioral change These long term result s can be compared to those of Lutes and colleagues (2008) and Sbrocco and colleagues (1999), in which participants who were assigned greater behavioral changes experienced larger amounts of weight regain. Similar results of weight regain were found in a r andomized trial where participants were categorized into tertiles according to maximum amount of weight loss (Jeffery et al., 1998). Participants who achieved the greatest amount of weight reduction in the short term also experienced a larger and more rap id weight regain than those who initially demonstrated slower, smaller losses. Weiss and colleagues (2007) also found in an analysis of the 1999 2002 National Health and Nutrition Examination Survey that losing a greater percentage of maximum weight was a ssociated with overall greater weight regain. While this may have been true in the present study, the initial behavior change provided great er benefit a s is illustrated by the significantly greater proportion of participants in the 1,000 kcal/day conditi on achieving and maintaining a > 5% weight reduction at 12 months. In the aforementioned trial by Jeffery and colleagues (1998), participants who lost more weight initially also regained more weight; however, they continued to experience a greater reducti on at 30 months compared to those who initially los t less. Unlike previous research that utilized the small change approach (Lutes et al., 2008; Sbrocco et al., 1999) where participants who made small behavioral changes continued to lose weight, participa
76 condition maintained, but did not continue to lose weight. This, along with the fact that less than half of the participants in this condition achieved a > 5% reduction, suggests that the changes may not have been large enough to achieve beneficial results. Therefore, prescribing larger behavioral changes m ay still provide more benefit s regardless of the greater weight regain. This study included an extended care program which has be en found to be one of the mo st successful strategies utilized to promote maintenance of learned behaviors as well as lost weight ( Perri and Corsica, 2002; Vetter et al., 2010 ). P articipants in the present study received extended care in the form of group meetings once per month. Wh ile it was posited that this extended care would promote continued use of cognitive behavioral techniques (e.g., goal setting, self monitoring, and problem solving), g iven the difficulties with attendance and adherence rates, it is questi onable whether thi s approach served as the best method. Another form of extended care as used in the TOURS project (Perri et al., 2008), provided participants with follow up care twice per month in the year following initial treatment and found lower levels of weight rega in. Others have utilized a step down or tapered approach from treatment where participants receive sessions twice per month for three months follow ed by monthly sessions (Wadden et al., 2010). Perhaps th e s e alternate methods would have provided additiona l group support continued focus on cognitive behavioral techniques accountability, and reinforcement to result in lower weight regain for participants in the present study who were asked to maintain large behavioral changes.
77 Secondary Aims With regard s to our secondary aim, the majority of participants in both the 1,000 and 1,500 kcal/day treatment conditions were found to be consuming within (+/ ) 15% of their prescribed kcal /day amount ( 53 % and 77 % of participants respectively) using classification from the PREFER behavioral treatment weight loss study (Warziski et al., 2008). Unlike the two previous studies where participants were prospectively prescribed caloric intake goals (Das et al., 2008; Toplak, et al., 2005), participants in the present stu dy consumed calorie amounts that were significantly different from each other during the first six months of the study with the 1,000 kcal/day condition consuming a mean 1,164 + 170 calories and the 1,500 kcal/day condition consuming a mean 1,518 + 222 ca lories These results could be attributed in part methodology that utilized a lifestyle intervention with components of self monitoring, structured meal plans, and suggested meal replacements and portion controlled meals. Following active treatment, caloric intake between treatment condition s remained significantly different by a margin of 24 2 calories. However, given the decreased rates of self monitoring, it remains questionable whether this serves as an accurate measurement of actual c aloric intake for the treatment conditions. Conversely, two previous studies that prospectively prescribed energy intake amounts found that participants tended to gravitate toward a median dietary intake they find manageable regardless of energy prescripti on (Das et al., 2008; Toplak et al., 2005). Toplak and colleagues (2005) prescribed dietary restriction in combination with medication within a medical setting. While the investigators did provide dietary counseling, participants did not receive a behav ioral lifestyle intervention where
78 techniques to modify eating habits were addressed. In addition, Das and colleagues (2008) provided food for 24 weeks and then asked participants to continue eating either a mild or moderate calorie restricted diet for an additional 24 weeks. The participants did meet throughout the study for support groups individual nutritional counseling with a di etician and safety monitoring ; however, a lifestyle intervention utilizing behavioral techniques was not provided. Resu lts of the present study indicated however, that the treatment conditions did differ in adherence to their respective calorie intake goals. While the mild caloric restriction condition was found to consume only 1 % greater calories than their 1 ,500 kcal/d ay goal participants within the 1,000 kcal/day condition consumed a mean 16% calories above their prescribed amount Some 45% of the participants in the 1,000 kcal/day condition consumed > 15% above their calorie goal. These results may provide addition al support for the small change approach (Hill et al., 2003) which suggests that mildly reducing calories to 1,500 kcal/day may be more feasible than moderately reducing calories to 1,000 kcal/day. C (1980) disinhibition hyp othesis, restraining eating as low as 1,000 kcal/day may lead to all or nothing thinking, and violations to dietary goals may r esult in overconsumption Tertiary and Exploratory Aims This study also assessed the i mpact of the 1,000 and 1,5 00 caloric presc riptions on reported levels of self efficacy, dietary disinhibition, restraint, hunger, and depression Results indicated that the two conditions did not differ at baseline, Month 6, or Month 12 on any of these factors. Trends for both conditions were co nsistent with previous findings (
79 Warziski et al., 2008; Wing et al., 2008) With regards to self efficacy, women in the present study were found to significantly increase levels of self effic acy from baseline to long term follow up. Levels of self efficacy were a ssociated with weight loss at 12 months, with those women displaying the highest levels of self efficacy, regardless of treatment condition, achieving the greatest body weight reducti on. These res ults are consistent with those of Warziski and colleagues (2008) who demonstrated that, with the use of a behavioral intervention, self efficacy can be increased over the course of treatment and is associated with long term weight management. Wing and colleagues (2008) found that increases in disinhibition and hunger as well as decreases in restraint were related to greater weight regain ; h owever, this was only found in participants who did not receive in person extended care follow up. The present study d id utilize a six month extended care period, in which participants were encouraged to maintain use of strategies to manage these behaviors. Therefore, p articipants reported significant increases in levels of restraint and lower levels of d isinhibition across the study period, a trend that appears similar to those found between these 1985; Wadden et al., 1993). Given both treatment conditions in the present study experi enced overall long term weight loss, this suggests that participants learned to control periods of overindulgence through increased dietary self control (Foster et al., 1998). A discrepancy did exist, however, as it was found during Phase II that those pa rticipants who regained > 4.5 kg reported greater disinhibition and hunger and decreased restraint regardless of treatment condition Such changes in these variables have previously been linked to weight regain (Hays & Roberts, 2008; Wing et al., 2008).
80 Susceptibility to eat in response to hunger has been linked to decreased restraint and increased disinhibition (Stunkard & Messick, 1985). Participants within both treatment conditions endorsed reductions in levels of hunger over the course of treatmen t, with no differences existing between condition s. Previous findings of severe caloric restriction found that increased levels of hunger were associated with greater caloric intake (LaPorte & Stunkard, 1990). Participants in this study were taught strat egies to manage their cravings and learned ways to consume a balanced diet to reduce levels of hunger. Therefore, these learned techniques may have counteracted the potential for increased hunger, specifically for participants in the moderate caloric rest riction 1,000 kcal/day condition who may have been more likely to report greater hunger. At baseline, participants within both condition s on average, endorsed only minimal levels of depression and did not fall within the classification of binge eaters. Throughout the course of treatment, levels of these two variables continued to decrease It has previously been found that increases in reported depressive symptoms and uncontrolled eating tendencies may result in difficulties maintaining weight loss (Win g et al., 2008) Similar results were found in this study in those who regained > 4.5 kg Specifically, d uring Phase II, it was found that participants who regained a larger amount of weight reported greater uncontrolled eating and higher levels of depr ession, regardless of treatment condition, than those who regained < 4.5 kg Taken together, it is therefore suggested that, using lifestyle interventions with a period of extended care like that used in the current trial may teach the majority of parti cipants strategies to increase self efficacy and restraint while lowering levels of disinhibition, hunger, depression, and episodes of binge eating. T hese results were
81 found regardless of prescribed caloric intake goal Perhaps this is due to the fact th at participants received the same intervention and learn ed similar strategies. Cognitive behavioral techniques to modify eating and activity patterns teach participants methods not only to produce a negative energy balance necessary to lose weight, but a lso ways to identify and modify antecedents and consequences associated with unhealthy behavioral patterns as well as ways to modify the environment to develop healthy behavioral habits (Vettner et al., 2010; Ferster et al., 1962; Stuart, 1967; Wing, 2002) In line with the concepts of the social cognitive theory (Bandura, 1977, 1986; Wadden & Foster, 1992), lifestyle interventions not only increase knowledge of health behaviors, these efficacy. behavior, cognitions, and environment using behavioral techniques of pro blem solving, goal setting, self monitoring, self reinforcement, stimulus control, and cognitive restructuring (Bandura, 1977, 1986; Wadden & Foster, 1992). Because participants in both caloric prescription conditions received the same lifestyle intervent ion, all learned these behaviors. Therefore, this structure may explain why, regardless of condition assignment, t rends showed reductions in disinhibition, hunger, and depression, and increases in restraint. The present study also found overall increases in six out of nine factors of quality of life. Across both treatment conditions, participants endorsed improvements in Health Transition, Physical Functioning, General Health, Vitality, Social Functioning, and Mental Health across the study period. Impr ovements in quality of life have previously
82 been linked to greater weight loss outcomes (Carels et al., 2003; Kolotkin et al., 2001) and may again result from the structure of the lifestyle intervention, which included a group format and utilized cognitive behavioral techniques to overcome eating and exercise barriers. There existed a d ifference between caloric prescription conditions with respect to reported levels of Social Functioning at Month 12 Because no other variables of quality of life or levels of depression, binge eating, disinhibition, restraint, hunger, or self efficacy varied between condition s, it remains questionable whether the difference in Social Functioning comes as a result of calor ic prescription, or if there exist other non study re lated variables that could account for this difference. In further exploring potential reasons for the large standard deviations at Month 12, it was found that 28 women regained a significant amount o f weight during Phase II. At baseline, these women r eported greater levels of depression disinhibition and caloric intake. The link among these variables and weight regain appears congruent with theory outlined by Polivy and Herman (1983). These researchers suggested that affective sta tes, such as depre ssion, may decrease motivation to diet and result in disinhibition of eating Similarly, a violation of intake goals (e.g. a restrained eater consuming excessive calories) may result in negative affect and an abandonment of eating patterns (Polivy & Herma n, 1983). Given the participants who regained the most weight displayed higher baseline levels of depression, dis inhibition, and caloric intake deviations from caloric prescription goals during the intervention or changes in affective states may have ins tigate d the habitual pattern of overeating and weight regain. N o interaction effects existed between depression or disinhibition and treatment condition with respect to weight regain during Phase II This suggests that regardless
83 of treatment placement, participants who exhibit greater de pression and disinhibition at baseline ma y perform worse in the long term. Treatment condition appeared to serve as a moderator between r eported baseline caloric intake and weight regain during Phase II Women who consu med more calories at baseline regained more weight if they were assigned to the 1,000 kcal/day group than those who were assigned to the 1,500 kcal/day group. This suggests that matching participants who initially consume a greater number of calories to a lower caloric intake prescription thus requiring greater reductions in caloric intake, may be detrimental to long term weight maintenance. The disinhibition hypothesis and restraint theory ( Polivy & Herman, 1983; Ruderman, 1986) suggest that restrained eaters often view dieting as an all or nothing event and the perception of having overeaten disinhibits these eaters Given those women who consumed the most baseline calories and who were assigned the 1,000 kcal/day goal participate d in the highest leve l of dietary restriction (compared to those assigned the 1,500 kcal/day goal) v iolations to dietary control may have led to abandonment of the calorie goal Indeed, following Phase I of treatment, it was found that participants in the 1,000 kcal/day cond ition increased their caloric intake, which suggests difficulty in long term adherence to the goal Perhaps these participants who also experienced greater weight regain may have benefitted from a calorie goal range (e.g. 1,000 to 1,250 kcal/day) rather t han attempt ing to maintain their restrictive goal from Phase I. In fitting with the restraint theory (Polivy & Herman, 1983), this may have reduced all or nothing perception of the dietary goal and lowered disinhibition of eating patterns and ensuing weig ht regain
84 During the intervention, women who regained > 4.5 kg attended a greater number of sessions compared to those who regained < 4.5 kg. Perhaps the participants who regained the most weight utilized the group support and accountability during Phase I of the intervention but when this was no longer available on a weekly basis, were unable to maintain their habit changes Stunkard and Messnick (1985) posited that persons who score high on baseline disinhibition may respond well to the support associ ated with group approaches, especially when comorbid depression serve s as an emotional disinhibitor. As these participants moved to Phase II, t he monthly extended car e sessions may have been too few, the 1,000 kcal/day goal may have been too difficult to maintain, and the lack of group support may have made it hard to continue the behavioral changes necessary to maintain weight loss. Indeed, during Phase II, participants who regained > 4.5 kg maintained significantly fewer self monitoring records than tho se who regained < 4.5 kg, a behavior that has been highly correlated with lower caloric intake and greater long term weight loss (Baker & Kirschenbaum, 1993; Boutelle & Kirschenbaum, 1998). In addition the large regainers reported significantly lower lev els of restraint and higher levels of depression, disinhibition, uncontrolled eating, and hunger at Month 12 compared to those participants who regained < 4.5 kg. These results mirror those of Wing and colleagues (2008), who found that greater disinhibiti on, uncontrolled eating, and hunger as well as lower restraint were related to greater weight regain. In the group who regained > 4.5 kg, levels of self efficacy were also found to be significantly lower than those who did not regain that magnitude of wei ght. While higher self efficacy has been associated with greater weight loss success (Richman et al., 2001; Warziski
85 et al., 2008), poorer self efficacy has been linked to weight cycling (Kensinger et al., 1998). Bandura (1986) defined perceived self eff icacy as the estimation of capabilities t o perform a behavior. As participants who regained larger amounts of weight lost weekly group support and accountability, they may have experienced a decreased cap ability to maintain the healthy behaviors learned d uring Phase I. Limitations There are three potential limitations to the present study. First, participants self reported caloric intake. When self reporting energy intake, research suggests that people tend to underestimate consumption by 27 to 46% (Joh nson et al., 2005; McKenzie et al., present herself in a p ositive, acceptable manner (McKenzie et al., 2002; Mertz, Tsui, Judd, et al., 1991) or from forgetfulness due to the inconvenie nce of record keeping (Macdiarmid & Blundell, 1997; Price, Paul, Cole, & Wadsworth, 1997). As previous research suggests, our sample of obese women tend ed to underreport caloric intake at a higher prevalence than other groups (Johnson, Go ran, & Poehlman, 1994; Lichtman et al., 1992). Due to study constraints, utilization of doubly labeled water, which is often seen as the gold standard technique to measure energy expenditure and, in turn, energy intake (Livingstone, & Black, 2003 ; Trabulsi, & Schoeller, 2001 ) was not feasible to technically address the question of caloric intake Second, attendance and adherence rates specifically during Phase II of the present study were lower than those reported in other trials involving a lifestyle intervention and extended care (Perri et al., 2008; Wadden et al., 2009). Unlike the aforementioned studies, women in the present study were not provided factors of extrinsic motivation for participation or attendance, such as monetary incentives or meal
86 replacements which may have impacted the rates of attendance Given session attendanc e (Carels et al., 2003; Wadden et al., 1992) and adherence to self monitoring records (Baker & Kirschenbaum, 1993; Boutelle & Kirschenbaum, 1998) have been associated with amount of long term weight loss weight reduction results and amounts of regain may have been adversely impacted due to low rates of attendance and adherence Finally, the study is also limited by the population assessed; namely, only women between the ages of 25 an d 75 weighing between 91 to 136 kg (i.e. 200 to 300 lbs), and having a BMI between 30 and 45 kg/m 2 were included in our sample. Generalizability therefore is limited by the exclusion of men and individuals of various weight and BMI groups. Additionally, those with serious health conditions who were unable to obtain permission from their physician were excluded from study participation, causing a truncated range that excludes individuals with uncontrolled comorbid medical conditions Clinical Implicatio ns The present study adds important clinical implications with regards to behavioral treatments for weight loss. While traditional behavioral treatment programs for obesity often result in a reduction of body weight that is associated with clinically sig nificant benefits in health challenge in the long There exist two competing beliefs regarding prevention of weight regain, with some arguing fo r a small changes approach that will lead to slow, yet steady reductions in body weight and attainable and sustainable healthy behavior changes (Hill, 2009; Hill, Wyatt, Reed &
87 Peters, 2003). Conversely, others support fast initial behavior changes and la rger initial weight losses for greater long term weight reduction (Astrup & Rssner, 2000; Carels et al., 2003; Elfhag & Rssner, 2005; Elfhag & Rssner, 2010; Jeffery et al., 1998; Nackers et al., Wadden, et al., 1992). Results from the current study fo und that women prescribed a moderate caloric restriction of 1,000 kcal/day lost more weight initially, but experienced greater regain so that weight change did not differ from that of participants prescribed a 1,500 kcal/day goal at 12 month follow up Th is appears to be in accordance with the notion that larger initial weight loss are associated with greater long term weight regain (Jeffery et al., 1998; Sbrocco et al., 1999; Wadden, Foster, & Letizia, 1994), suggesting larger initial weight reduction may serve as a risk factor for later weight regain (McGuire, Wing, Klem, Lang, & Hill, 1999; Weiss et al., 2007; Wing & Hill, 2001). This also rai ses the question of participant s to maintain large changes in diet and exercise necessary to prevent s usage of a six month extended care period. that suggests that large behavioral changes early in treatment may initially provi de enough reinforcement value (i.e. fast weight loss) to result in short term habit change; however, they may als o be difficult to maintain long term, thus resulting in larger weight regain. Clinically, these res ults also suggest that extended care progra ms may need to be modified to continually reinforce healthy behavior changes achieved during the initial months of a lifestyle intervention program. The present study did find, however, that participants were able to adhere to their caloric prescriptions during active trea tment, thus
88 suggesting that self monitoring, structured meal plans, and suggested meal replacements and portion controlled meals are useful methods to increase adherence. Increased emphasis has been placed on clinical versus statistical differences. At Month 12, women in the 1,000 kcal/day condition maintained a 2.8 kg larger weight reduction than those in the 1,500 kcal/day condition In addition, 6 1 % of participants in the 1,000 kcal/day condition compared to 42 % of those in the 1,500 kcal/day condition achieved a beneficial 5% weight reduction at 12 month. This study may not have been adequately powered to detect a significant difference at this time point. Therefore, while not statistically different, for clinical purposes of achie ving an overall 2.8 kg larger weight loss and a greater likelihood of reaching a > 5% weight change a moderate caloric restriction goal of 1,000 kcal/day may be superior. Encouraging participants to make larger changes early on to result in greater initi al weight loss may increase the likelihood of achieving a > 5% reduction long term. This remains in accordance with the context of lifestyle interventions where cognitive behavioral strategies are taught to safely achieve and maintain health y eating and exercise behaviors and where food intake remains above 1,000 kcal/day Additional long term follow up of participants who are prescribed various levels of caloric intake are warranted. As participants in the 1,000 kcal/day condition were more likely t o achieve a > 5% weight reduction, they also experienced greater weight regain from Months 6 12 It remains unclear whether this rate of regain would continue so that condition s no longer differ ed in the proportion of participants achieving a > 5% weigh t reduction. Future work should also provide special consideration to participants who endorse greater symptoms of depression, higher levels of disinhibition, and larger
89 caloric intakes at baseline T hese baseline indicators were reported by participants in the present study who regained large amounts of weight in the long term Treatment matching may prove an important strategy, specifically with participants reporting greater baseline caloric intake Findings from this study demonstrated that prescri bing participants who consumed more calories at baseline to the 1,000 kcal/day resulted in larger long term weight regain than prescribing these participants to a 1,500 kcal/day goal Therefore, matching participants with higher baseline calorie levels to a milder and more gradual caloric reduction may improve long term weight loss outcomes. Another potential option of staging or providing a range of maintenance goals may reduce weight regain, specifically for those assigned to a 1,000 kcal/day goal who c onsume higher calorie levels prior to treatment. Rather than encouraging maintenance of a 1,000 kcal/day intake during extended care unless a normal BMI has been achieved, gradually moving partic ipants from 1,000 to 1,250 to 1,500 kcal/day goal s or allowi ng a range of 1,000 to 1,250 kcal/day may increase levels of self efficacy and reduce all or nothing attitudes that could lead to abandonment of diet ( Polivy & Herman, 1983; Ruderman, 1986). In addition, studies assessing various fr equencies and lengths o f extend care periods may discover whether participants assigned to various caloric restriction goals respond differently to the cognitive behavioral strategies required to maintain weight loss long term. Future s tudies are warranted to assess whether add itional personal and behavioral baseline characteristics not measured in this study predict who will adhere to and achieve success with a caloric restriction goal of 1,000 kcal/day
90 In sum, findings sugge st that, within the context of lifestyle treatment prescribing a g reater caloric restriction goal of 1,000 kcal/day may prove beneficial for initial short term weight loss, but the behavioral changes required to sustain large initial changes may be too great for long term maintenance when compared to a m ild caloric restriction goal of 1,500 kcal/day However, while not statistically different, at long term follow up those prescribed a 1,000 kcal/day goal continued to demonstrate a clinically larger weight loss. A significantly greater percentage of part icipants within this condition also achieved a beneficial weight reduction > 5% than those in the 1,500 kcal/day condition. Given the potential health benefits associated with weight losses of this magnitude, these results suggest important advantages to prescribing a 1,000 kcal/day intake goal during lifestyle interventions Regardless of caloric prescription, lifestyle interventions serve as important avenues to increase restraint and self efficacy, and decrease disinhibition, depression, binge eating, and hunger. Emphasi zing attendance and adherence to self monitoring records also plays a role in achieving gre ater long term weight loss Finally, as treatment condition appeared to moderate the association between baseline caloric intake and weight rega in, matching those participants with higher baseline caloric intake to more mild caloric restriction goals, or providing tailored extended care may result in greater long term weight management success.
91 APPENDIX A SAMPLE METHODS TO MEET CALORIE INTA KE GOAL SAMPLE MEAL PLANS: 1,000 CALORIES Sample Meal Plan Sample Meal Plan 2 Calories Calories Breakfast: Breakfast: 1 cup sliced strawberries 53 1 medium orange 62 1 tsp. sugar substitute 0 1/4 cup egg substitute 53 2 slices light wh ole wheat toast 120 2 slices onion, chopped 15 2 tsp. light margarine or butter 28 1/4 cup bell pepper 10 6 oz. fat free, sugar free yogurt 100 1 slice light whole wheat toast 60 Total: 301 1 tsp. peanut butter 31 Total: 231 AM Snack : 10 fresh grapes 35 AM Snack : 1 small apple 60 Lunch: 2 oz. 98% fat free sliced turkey breast 60 Lunch: 1/2 cup fresh spinach or romaine lettuce 12 2 oz. skinless, grilled chicken breast 94 2 tbsp. shredded parmesan cheese 41 2 c ups salad greens (not iceberg) 18 2 tsp. yellow or dijon mustard 2 1 tbsp. reduced fat feta cheese 35 1 whole wheat english muffin 150 1 tsp. olive oil 39 10 baby carrots, raw 35 1 tbsp. red wine vinegar 8 2 tbsp. fat free ranch dressing 30 4 salti ne crackers 48 Total: 330 1 peach 60 Total: 302 PM Snack : 1/2 medium banana 53 PM Snack : 6 oz. fat free, sugar free yogurt 100 Dinner : 4 oz. baked salmon filet 100 Dinner : 1 medium baked sweet potato 130 3 oz. large shrimp, cooked w/o fat 131 1/2 tsp. light margarine 8 1 cup broccoli, steamed 20 1 cup steamed green beans 45 1/2 cup red bell pepper 20 Total: 283 1/2 cup summer squash 11 1/2 tsp. olive oil 19 Total for Day: 1002 1/3 cup brown rice, cooked 111 Total: 312 Total for Day: 1005
92 S AMPLE MEAL PLANS: 1,500 CALORIES Sample Meal Plan Sample Meal Plan 2 Calories Calories Breakfast: Breakfast: 1 cup oatmeal 146 1/3 cup egg substitute 71 cup fat free skim milk 45 cup green pepper 10 2 tbsp. raisins 85 2 slices onion, chopped 15 1 slice whole wheat toast 60 1 tsp. olive oil 39 1 tsp. peanut butter 31 1 slice light whole wheat toast 60 Total: 367 1 tsp. peanut butter 31 1 medium banana 105 AM Snac k: Total: 331 1 oz. plain almonds 164 AM Snack: Lunch: 2 cups fresh strawberries 106 4 oz. tuna in water 132 6 oz. fat free, sugar free yogurt 100 2 slices whole grain bread 120 206 2 tsp. yellow or Dijon mustard 2 Lunch: 2 s lices fresh tomato 5 2 cups fresh spinach 48 cup fresh spinach or romaine 12 2 oz. skinless grilled chicken 94 10 baby carrots, raw 35 1 tbsp. reduced fat feta cheese 70 2 tbsp. fat free ranch dressing 30 1 tsp. olive oil 39 Total: 336 1 tbsp. r ed wine vinegar 8 1 whole wheat roll (1 oz.) 75 PM Snack: 1 cup nonfat skim milk 91 6 oz. fat free, sugar free yogurt 100 Total: 425 cup sliced strawberries 26 126 PM Snack: Dinner: 1 oz. plain almonds 164 4 oz. pork tenderloi n 185 1/3 cup brown rice 111 Dinner: 1/2 cup beets 37 4 oz. large shrimp, cooked 197 cup zucchini 14 1 cup broccoli, steamed 20 1 cup mixed salad greens 18 1/2 cup red bell pepper 20 2 tsp. olive oil 40 1/2 cup summer squash 11 1 tbsp. red wine or other vinegar 8 1/2 tsp. olive oil 19 1 cup fat free skim milk 91 1/3 cup brown rice, cooked 111 Total: 504 Total: 378 Total for Day: 1497 Total for Day: 1504
93 EXAMPLES OF PORTION CONTROLLED FOODS Lean Cuisine Calories Alfredo Pasta with Chicken & Broccoli 300 Angel Hair Pomodoro 250 Asian Style Pot Stickers 260 Baja Style Chicken Quesadilla 280 BBQ Chicken Quesadilla 260 Cheddar Potatoes with Broccoli 230
94 EXAMPLES OF PARTIAL MEAL REPLACEMENTS Slim Fast Calories Slim Fast Creamy Milk Chocolate Shake 190 Slim Fast Chocolate Cookie Dough Meal Bar 220 EXAMPLE OF PORTION CO NTROLLED EATING PLAN: 1,000 CALORIES Calories Breakfast: Slim Fast Milk Shake 190 1 medium banana 105 Total: 2 95 Lunch: Slim Fast Meal Bar 220 Dinner: Lean Cuisine Asian Style Pot Stickers 250 1 whole wheat roll (1 oz) 75 1/2 cup beets 37 cup zucchini 14 1 cup mixed salad greens 18 1 tsp. olive oil 20 1 tbsp. red wine or other vinegar 8 1 small apple 60 Total: 482 Total for Day: 997
95 APPENDIX B GROUP PERFERENCE QUESTIONNAIRE Please answer the following questions b ased on your preferences: 1. When you think about reducing the amount you eat, which would you prefer ? experience a rapid rate of weight loss, but eat fewer calories per day. experience a slower rate of weight loss, but eat more calories per day. 2. When you think about meeting in a group that discusses weight, nut rition, and physical activity, which would you prefer ? would be expected to share my personal experiences and progress related to making changes in my diet and physical activity. which shar ing my personal experiences and progress related to making changes in my diet and physical activity would be optional.
96 LIST OF REFERENCES Alley, D. E., & Chang, V. W. (2 007). The changing relationship of obesity and disability, 1988 2004. Journal of the American Medical Association, 298 2020 2027. American College of Sports Medicine. (2001). Appropriate intervention strategies for weight loss and prevention of weight re gain in adults. Medicine & Science in Sports & Exercise, 33, 2154 2156. Ashley, J. M., Herzog, H., Clodfelter, S., Bovee, V., Schrage, J., & Pritsos, C. (2007). Nutrient adequacy during weight loss interventions: a randomized study in women comparing t he dietary intake in a meal replacement group with a traditional food group. Nutrition Journal, 6 12 Astrup, A., & Rssner, S. (2000). Lessons from obesity management programmes: greater initial weight loss improves long term maintenance. Obesity Review s, 1 17 19 Ayyad, C., Andersen, T. (2000). Long term efficacy of dietary treatment of obesity: a systematic review of studies published between 1931 and 1999. Obesity Reviews, 1 113 119. Baker, R. C. & Kirschenbaum, D. S. (1993). Self monitoring may b e necessary for successful weight control. Behavior Therapy, 24 377 394. Bandura, A. (1977). Social learning theory. Englewood Cliffs NJ : Prentice Hall. Bandura. A. (1986). Social foundations of thought and action: a social cognitive theory. Englewood Cliffs NJ : Prentice Hall. Baron, R. M., & Kenny, D. A. (1986). The moderator mediator variable distinction in social psychological research: conceptual strategic and statistical considerations. Journal of Personality and Social Psychology, 51 1173 1182 Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory II San Antonio, TX: Psychological Corporation. Blair, S. N., & Leermakers, E. A. (2002). Exercise and weight management. In T. A. Wadden & A. J. Stunkard (Eds. ), Handbook of obesity treatment (pp. 283 300). New York NY : Guilford Press. Blundell, J. E. (2000). What foods do people habitually eat? A dilemma for nutrition, an enigma for psychology. American Journal of Clinical Nutrition, 71 3 5.
97 Brown, C. D., H iggins, M., Donato, K. A., Rohde, F. C., Garrison, R., Obarzanek, E., et al. (2000). Body mass index and the prevalence of hypertension and dyslipidemia. Obesity Research, 8 605 619. Brownell, K. D. (2000). The LEARN program for weight management 2000 D allas, TX: American Health Publishing Company. Bruce, B., & Wilfley, D. (1996). Binge eating among the overweight population: a serious and prevalent problem. Journal of the American Dietetic Association, 96 58 61. Boutelle, K. N., & Kirschenbaum, D. S. (1998). Further support for consistent self monitoring as a vital component of successful weight control. Obesity Research, 6 219 224. Burke, L. E., Warziski, M., Styn, M. A., Music, E., Hudson, A. G., & Sereika, S. M. (2008). A randomized clinical tria l of a standard versus vegetarian diet for weight loss: the impact of treatment preference. International Journal of Obesity, 32 166 176. Carels, R. A., Cacciapaglia, H. M., Douglass, O. H. (2003). The early identification of program. Eating Behaviors, 4 265 282. Clark, M. M., Abrams, D. B., Niaura, R. S., Eaton, C. A., & Rossi, J. S. (1991). Self efficacy in weight management. Journal of Consulting and Clinical Psychology, 59 739 744. Carpenter, K. M., Hasin, D. S., Allison, D. B., & Faith, M. S. (2000). Relationship between obesity and DSM IV major depressive disorder, suicide ideation, and suicide attempts: results from a general population study. American Journal of Public Health, 90 251 257. Chang, V. W., & Lauderdale, D. S. (2005). Income disparities in body mass index and obesity in the United States, 1971 2002. Archives of Internal Medicine, 165 2122 2128. Craig, C. L., Marshall, A. L., Sjstrm, M., Bauman, A. E., Booth, M. L., Ainsworth, B. E., et al. (2003). International physical activity questionnaire: 12 country reliability and validity. Medicine & Science in Sports & Exercise, 35 1381 1395. & Scucchi, S. (2001). Relationship between dietary restraint, binge eating, and leptin in obese women. International Journal of Obesity, 25 373 377.
98 Das, S. K. Saltzman, E., Gilhooly, C. H., DeLany, J. P., Golden, J. K., Pittas, A. G., et al. (2009). Lo w or moderate dietary energy restriction for long term weight loss: what works best? Obesity 17 2019 2024. Debraganza, N. (2010). The contributions of frequency, consistency, and comprehensiveness of self monitoring food records to weight regain. Doctor al dissertation, University of Florida, 2010. Diabetes Prevention Program Research Group. (2002). Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. New England Journal of Medicine, 346 393 403 Donnelly, J. E., Blair S. N., Jakicic, J. M., Manore, M. M., Rankin, J. W., & Smith, B. K. (2009). Appropriate physical activity intervention strategies for weight loss and prevention of weight regain for adults. Medicine & Science in Sports & Exercise, 41 459 471. Dumitresc u, R. G., & Cotarla, I. (2005). Understanding breast cancer risk Where do we stand in 2005? Journal of Cellular and Molecular Medicine 9, 208 221. Egger, G. J. (2006). Are meal replacements an effective clinical tool for weight loss? Medical Journal of Australia, 184 52 53. Elfhag, K., & Rssner, S. (2010). Initial weight loss is the best predictor for success in obesity treatment and sociodemographic liabilities increase risk for drop out. Patient Education and Counseling, 79 361 366. Elfhag, K., & Rssner, S. (2005). Who succeeds in maintaining weight loss? A conceptual review of factors associated with weight loss maintenance and weight regain. Obesity Reviews, 6 67 85. Engel, S. G., Crosby, R. D., Kolotkin, R. L., Hartley, G. G., Williams, G. R., Wonderlich, S. A., et al. (2003). The impact of weight loss and regain on obesity specific quality of life: mirror image or differential effect. Obesity Research, 11 1207 1213. Ferster, C. B., Nurnberger, J. I., & Levitt, E. E. (1962). The control o f eating. Journal of Mathetics, 1 87 109. Field, A. E., Coakley, E. H., Must, A., Spadano, J. L., Laird, N., Dietz, W. H., et al. (2001). Impact of overweight on the risk of developing common chronic disease during a 10 year period. Archives of Internal Medicine, 161 1581 1586. Finkelstein, E. A., Fiebelkorn, I. C., & Wang, G. (2003). National medical spending Health Affairs W3, 219 226.
99 Finkelstein, E. A., Trogdon, J. G., Cohen, J. W, & Dietz, W. (2009). Annual medical spending attributable to obesity: payer and service specific estimates. Health Affairs, 28 w822 w831. Flegal, K. M., Carroll, M. D., Kuczmarski, R. J., & Johnson, C. L. (1998). Overweight and obesity in the United S tates: Prevalence and trends, 1960 1994. International Journal of Obesity, 22 39 47. Flegal, K. M., Carroll, M. D., Ogden, C. L., & Curtin, L. R. (2010). Prevalence and trends in obesity among US adults, 1999 2008. Journal of the American Medical Associa tion, 303 235 241. Flegal, K. M., Carroll, M. D., Ogden, C. L., & Johnson, C. L. (2002). Prevalence and trends in obesity among US adults, 1999 2000. Journal of the American Medical Association, 288 1723 1727. Flegal, K. M., Graubard, B. I., Williamso n, D. F., & Gail, M. H. (2005). Excess deaths associated with underweight, overweight, and obesity. Journal of the American Medical Association 293, 1861 1867. Flegal, K. M., Graubard, B. I., Williamson, D. F., & Gain, M. H. (2007). Cause specific exces s deaths associated with underweight, overweight, and obesity. Journal of the American Medical Association, 298 2028 2037. Fogelholm, M., Kukkonen Harjula, K., & Oja, P. (1999). Eating control and physical activity as determinants of short term weight ma intenance after a very low calorie diet among obese women. International Journal of Obesity, 23 203 210. Fontaine, K. R., & Barofsky, I. (2001). Obesity and health related quality of life. Obesity Reviews 2 173 182. Foster, G. D., Wadden, T. A., Swain, R. M., Stunkard, A. J., Platte, P., & Vogt, R. A. (1998). The Eating Inventory in obese women: clinical correlates and relationship to weight loss. International Journal of Obesity and Related Metabolic Disorders, 22 778 785. Freedland, S. J., & Aronson W. J. (2005). Obesity and prostate cancer. Urology 65, 433 439. Gormally, J., Black, S., Daston, S., & Rardin, D. (1982). The assessment of binge eating severity among obese persons. Addictive Behaviors, 7 47 55. Gorman, B. S., & Allison, D. B. ( 1995). Measures of restrained eating. In D. B. Allison (Ed.), Handbook of assessment for eating behaviors and weight related problems. (pp. 149 184). Thousand Oaks, CA: Sage Publishing, Inc.
100 Greeno, C. G., Marcus, M. D., & Wing, R. R. (1995). Diagnosis of binge eating disorder: discrepancies between a questionnaire and clinical interview. International Journal of Eating Disorders, 17 153 160. Gregg, E. W., Cheng, Y. J., Cadwell, B. L., Imperatore, G., Williams, D. E., Flegal, K. M., et al. (2005). Secul ar trends in cardiovascular disease risk factors according to body mass index in US adults. Journal of the American Medical Association 293, 1868 1874. Hannum, S. M., Carson, L., Evans, E. M., Canene, K. A., Petr, E. L., Bui, L., et al. (2004). Use of portion controlled entrees enhances weight loss in women. Obesity Research, 12 539 546. Hays, N. P., & Roberts, S. Obesity, 16 52 58. H eatherton, T. F., Herman, C. P., Polivy, J., King, G. A., & McGree, S. T. (1988). The (mis)measurement of restraint: an analysis of conceptual and psychometric issues. Journal of Abnormal Psychology, 97 19 28. Heatherton, T. F., Polivy, J., & Herman, C. P. (1991). Restraint, weight loss, and variability of body weight. Journal of Abnormal Psychology, 100 78 83. Hedley, A. A., Ogden, C. L., Johnson, C. L., Carroll, M. D., Curtin, L. R., & Flegal, K. M. (2004). Prevalence of overweight and obesity among US children, adolescents and adults, 1999 2002. Journal of the American Medical Association, 291 2847 2850. Herman, C. P., & Mack, D. (1975). Restrained and unrestrained eating. Journal of Personality, 43 647 660. Herman, C. P., & Polivy, J. (1980). R estrained eating. In A. J. Stunkard (Ed.), Obesity (pp. 208 225). Philadelphia, PA: W. B. Saunders. Heymsfield, S. B., van Mierlo, C. A. J., van der Knaap, H. C. M., Heo, M., & Frier, H. I. (2003). Weight management using a meal replacement strategy: met a and pooling analysis from six studies. International Journal of Obesity, 27 537 549. Hill, J. O. (2009). Can a small changes approach help address the obesity epidemic? A report of the Joint Task Force of the American Society for Nutrition, Institute of Food Technologists, and International Food Information Council. American Journal of Clinical Nutrition, 89 477 484. Hill, J. O. Wyatt, H. R., Reed, G. W., & Peters, J. C. (2003). Obesity and the environment: where do we go from here? Science, 299 853 855.
101 Institute of Medicine (IOM). (1995). Weighing the options: criteria for evaluating weight management programs. Executive Summary Washington DC : National Academy Press. Jakicic, J. M., Clark, K., Coleman, E., Donnelly, J. E., Foreyt, J., Melanson, E., et al. (2001). Appropriate intervention strategies for weight loss and prevention of weight regain for adults. Medicine & Science in Sports & Exercise, 33 2145 2156. Jeffery, R. W., Drewnowski, A., Epstein, L. H., Stunkard, A. J., Wilson, G. T., Win g, R. R., & Hill, D. R. (2000). Long term maintenance of weight loss: current status. Health Psychology, 19 5 16. Jeffery, R. W., Wing, R. R., & Mayer, R. R. (1998). Are smaller weight losses or more achievable weight loss goals better in the long term f or obese patients? Journal of Consulting and Clinical Psychology, 66 641 645. Jeffery, R. W., Wing, R. R., Thorson, C., Burton, L. R., Raether, C., Harvey, J., & Mullen, M. (1993). Strengthening behavioral interventions for weight loss: a randomized tria l of food provision and monetary incentives. Journal of Consulting and Clinical Psychology, 61 1038 1045. Jenkinson, C., Wright, L., & Coulter, A. (1994). Criterion validity and reliability of the SF 36 in a population sample. Quality of Life Research, 3 7 12. Jia, H., & Lubetkin, E. I. (2005). The impact of obesity on health related quality of life in the general adult US population. Journal of Public Health, 27 156 164. Johnson, R. K., Friedman, A. B., Harvey Berino, J., Gold, B. C., & McKenzie, D. (2005). Underreporting among obese and overweight women. Journal of the American Dietetic Association, 105 1948 1951. Johnson, R. K., Goran, M., Poehlman, E. (1994). Correlates of over and underreporting of energy intake in healthy older men and women. A merican Journal of Clinical Nutrition, 59 1286 1290. Kanfer, F. H. (1970). Self monitoring: Methodological limitations and clinical applications. Journal of Consulting and Clinical Psychology, 35 148 152. Kensinger, G. J., Murtaugh, M. A., Reichmann, S K., & Tahgney, C. C. (1998). Psychological symptoms are greater among weight cycling women with severe binge eating behavior. Journal of the American Dietetic Association, 98 863 868. Keogh, J. B., & Clifton, P. M. (2005). The role of meal replacements in obesity treatment. Obesity Reviews, 6 229 234.
102 Kirschenbaum, D. S., & Tomarken, A. J. (1982). Some antecedents of regulatory eating in restrained and unrestrained eaters. Journal of Abnormal Psychology, 91, 326 336. Klesges, R. C., Isbell, T. R., & Klesges, L. M. (1992). Relationship between dietary restraint, energy intake, physical activity, and body weight: a prospective analysis. Journal of Abnormal Psychology, 101, 668 674. Kolotkin, R. L., Crosby, R. D., Williams, G. R., Hartley, G. G., & Nico l, S. (2001). The relationship between health related quality of life and weight loss. Obesity Research, 9 564 571. Kramer, F. M., Jeffery, R. W., Forster, J. L., & Snell, M. K. (1989). Long term follow up of behavioral treatment for obesity: patterns of weight gain among men and women. International Journal of Obesity, 13 124 136. Krebs, J. D., Evans, S., Cooney, L., Mishra, G. D., Fruhbeck, G., Finer, N., et al. (2002). Changes in risk factors for cardiovascular disease with body fat loss in obese wom en. Diabetes, Obesity and Metabolism, 4 379 387. LaPorte, D. J., & Stunkard, A. J. (1990). Predicting attrition and adherence to a very low calorie diet: a prospective investigation of the eating inventory. International Journal of Obesity, 14 197 206. LeCheminant, J. D., Jacobsen, D. J., Hall, M. A., & Donnelly, J. E. (2005). A comparison of meal replacements and medication in weight maintenance after weight loss. Journal of the American College of Nutrition, 24 347 353. Lichtman, S. W., Pisarska, K. Berman, E. R., Pestone, M., Dowling, H., Offenbacher, E., et al. (1992). Discrepancy between self reported and actual caloric intake and exercise in obese subjects. New England Journal of Medicine, 327 1892 1898. Linde, J. A., Rothman, A. J., Baldwin, A. S., & Jeffery, R. W. (2006). The impact of self efficacy on behavior change and weight change among overweight participants in a weight loss trial. Health and Psychology, 25 282 291. Little, R. J. (1994). A class of pattern mixture models for normal m issing data. Biometrika, 81 471 483. Livingstone, M. B., & Black, A. E. (2003). Markers of the validity of reported energy intake. Journal of Nutrition, 133 895S 920S. Lutes, L. D., Winett, R. A., Barger, S. D., Wojcik, J. R., Nichols Richardson, S., H erbert, W. G., & Anderson, E. (2008). Small changes in nutrition and physical activity promotes weight loss and maintenance: three month evidence from the ASPIRE randomized trial. Annals of Behavioral Medicine, 35 351 357.
103 Marcus, M. D., Wing, R. R., & La mparski, D. M. (1985). Binge eating and dietary restraint in obese patients. Addictive Behaviors, 10 163 168. Marlatt, G. A., & Gordon, J. R. (1980). Determinants of relapse: implications for the maintenance of behavior change. In P. O. Davidson & S. M. Davidson ( E ds.), Behavioral m edicine: changing health l ifestyles (pp. 410 452). New York NY : Brunner/Mazel. McGuire, M. T., Wing, R. R., Klem, M. L., Lang, W., & Hill, J. O. (1999). What predicts weight regain in a group of successful weight losers? Jour nal of Consulting and Clinical Psychology, 67 177 185. McKenzie, D. C., Johnson, R. K., Harvey Berino, J., & Gold, B. C. (2002). Impact of obese women. Obesity Research, 10 471 477. Melanson, K., & Dwyer, J. (2002). Popular diets for treatment of overweight and obesity In Wadden, T.A. & Stunkard, A.J. (Eds), Handbook of o besity t reatment (pp. 249 275) New York NY : Guilford Press. Mertz, W., Tsui, J. C., Judd, J. T., Reiser, S., Hallfrisch, J., Morris, E. R., Steele, P. D., Lashley, E. (1991). What are people really eating? The relation between energy intake derived from estimated diet records and intake determined to maintain body weight. American Journal of Clini cal Nutrition, 54 291 295. Mokdad, A. H., Ford, E. S., Bowman, B. A., Dietz, W. H., Vinicor, F., Bales, V. S., et al. (2003). Prevalence of obesity, diabetes, and obesity related health risk factors, 2001. Journal of the American Medical Association 2 89, 76 79. Mokdad, A. H., Marks, J. S., Stroup, D. F., & Gerberding, J. L. (2004). Actual causes of death in the United States, 2000. Journal of the American Medical Association, 291 (10), 1238 1245. Moore, L. L., Visioni, J. J., Wilson, P. o, R. B., Finkle, W. D., & Ellison, R. C. (2000). Can sustained weight loss in overweight individuals reduce the risk of diabetes mellitus? Epidemiology, 11 269 273. Nackers, L. M., Ross, K. M. & Perri, M. G. (2010). The impact of rate of initial weight loss on long term success in obesity treatment: does slow and steady win the race? International Journal of Behavioral Medicine, 17 161 167. National Cholesterol Education Program. (2001). Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP): Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Journal of the American Medical Association 285, 2486 2497.
104 National Heart, Lung and Blood Institute (NHLBI) (1998) Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: The evidence report. Obesity Research, 6 51S 209S. Niemeier, H. M. Phelan, S., Fava, J. L., & Wing, R. R. (2007). Internal disinhibition pr edicts weight regain following weight loss and weight loss maintenance. Obesity, 15 2485 2494. Noakes, M., Foster, P. R., Keogh, J. B., & Clifton, P. M. (2004). Meal replacements are as effective as structured weight loss diets for treating obesity in ad ults with features of metabolic syndrome. Journal of Nutrition, 134 1894 1899. Ogden, C. L., Carroll, M. D., Curtin, L. R., McDowell, M. A., Tabak, C. J., & Flegal, K. M. (2006). Prevalence of overweight and obe sity in the United States, 1999 2004. Journ al of the American Medical Association, 295 1549 1555. Patterson, P. D., Moore, C. G., Probst, J. C., & Shinogle, J. A. (2004). Obesity and physical inactivity in rural America. Journal of Rural Health, 20 151 159. Peeters, A., Barendregt, J. J., Wille kens, F., Mackenbach, J. P., Al Mamun, A., & Bonneux, L. (2003). Obesity in adulthood and its consequences for life expectancy: A life table analysis. Annals of Internal Medicine, 138 24 32. Perri, M. G. (1998). The maintenance of treatment effects in th e long term management of obesity. Clinical Psychology: Science and Practice, 5 526 543. Perri, M. G. & Corsica, J. A. (2002). Improving the maintenance of weight lost in behavioral treatment of obesity. In T.A. Wadden and A.J. Stunkard (Eds.), Handbook of obesity treatment (pp. 357 379). New York NY : Guilford Press. Perri, M. G., & Fuller, P. R. (1995). Success and failure in the treatment of obesity: Where do we go from here? Medicine, Exercise, Nutrition and Health, 4 255 272. Perri, M. G., Limach er, M. C., Durning, P. E., Janicke, D. M., Lutes, L. D., Bobroff, L. B., et al. (2008). Extended care programs for weight management in rural communities: the treatment of obesity in underserved rural settings (TOURS) randomized trial. Archives of Internal Medicine, 168 2347 2354. Polivy, J., Herman, C. P. (1976). Clinical depression and weight change: a complex relation. Journal of Abnormal Psychology, 85 338 340. Polivy, J., & Herman, C. P. (1983). Breaking the diet habit. New York NY : Basic Books. Polivy, J., & Herman, C. P. (1985). Dieting and binging. American Psychologist, 40 193 201.
105 Polivy, J., Herman, C. P., Hackett, R., & Kuleshnyk, I. (1986). The effects of self attention and public attention on eating in restrained and unrestrained subjec ts. Journal of Personality & Social Psychology, 50 1253 1260. Price, G. M., Paull, A. A., Cole, T. J., & Wadsworth, M. E. (1997). Characteristics of the low energy reporters in a longitudinal national dietary survey. British Journal of Nutrition, 77 833 851. Puhl, R. M., & Brownell, K. D. (2001). Bias, discrimination, and obesity. Obesity Research, 9 788 805. Puhl, R. M., Moss Racusin, C. A., Schwartz, M. B., & Brownell, K. D. (2008). Weight stigmatization and bias reduction: perspectives of overweigh t and obese adults. Health Education Research, 23 347 358. Renjilian, D. A., Perri, M. G., Nezu, A. M., McKelvey, W. F., Shermer, R. L., & Anton, S. D. (2001). Individual versus group therapy for obesity: effects of matching participants to their treatme nt preferences. Journal of Consulting and Clinical Psychology, 69 717 721. Richman, R. M., Loughnan, G. T., Droulers, A. M., Steinbeck, K. S., & Caterson, I. D. (2001). Self efficacy in relation to eating behaviour among obese and non obese women. Intern ational Journal of Obesity, 25 907 913. Rothacker, D. Q., Watemberg, S. (2004). Short term hunger intensity changes following ingestion of a meal replacement bar for weight control. International Journal of Food Science and Nutrition, 55 223 226. Ruder man, A. J. (1986). Dietary restraint: a theoretical and empirical review. Psychological Bulletin, 99 247 262. SAS Institute, Inc. (2008). SAS System, Version 9.2. Cary, NC: SAS Institute, Inc. Sbrocco, T., Nedegaard, R. C., Stone, J. M., & Lewis, E. L. (1999). Behavioral choice treatment promotes continuing weight loss: preliminary results of a cognitive behavioral decision based treatment for obesity. Journal of Consulting and Clinical Psychology, 67 260 266. Sbrocco, T., & Schlundt, D. G. (1998). App lying decision theory to understand cognitive Manuscript submitted for publication. Schlundt, D. G., Sbrocco, T., & Bell, C. (1989). Identification of high risk situations in a behavioral weight loss pro gram: application of the relapse prevention model. International Journal of Obesity, 13 223 234.
106 SPSS, Inc. (2010). SPSS for Windows, Rel. 18.0. Chicago, IL: SPSS, Inc. Stevens, V. J., Obarzanek, E., Cook, N. R., Lee, I., Appel, M. D., Smith West, D., e t al. (2001). Long term weight loss and changes in blood pressure: results of the Trials of Hypertension Prevention, Phase II. Annals of Internal Medicine, 134 1 11. Streit, K. J., Stevens, N. H., Stevens, V. J., & Rssner, J. (1991). Food records: A pre dictor and modifier of weight change in a long term weight loss program. Journal of the American Dietetic Association, 91 213 216. Stuart, R. B. (1967). Behavioral control of overeating. Behavioral Therapy, 5 357 365. Stunkard, A. J., & Messick, S (198 5). The three factor eating questionnaire to measure dietary restraint, disinhibition, and hunger. Journal of Psychometric Research, 29 71 83. Tataranni, P. A., & Ravussin, E. (2002). Energy metabolism and obesity. In T. A. Wadden & A. J. Stunkard ( E ds.) Handbook of obesity treatment (pp. 42 72) New York NY : Guilford Press. Teixeira, P. J., Going, S. B., Houtkooper, L B., Cussler, E. C., Metcalfe, L. L., Blew, R. M., Sardinha, L. B., Lohmann, T. G. (2004). Pretreatment predictors of attrition and su ccessful weight management in women. International Journal of Obesity and Related Metabolic Disorders, 28, 1124 11 33. Teixeira, P. J., Going, S. B., Sardinha, L. B., & Lohman, T. G. (2005). A review of psychosocial pre treatment predictors of weight cont rol. Obesity Reviews, 6 43 65. The Look AHEAD Research Group. (2010). Long term effects of a lifestyle intervention on weight and cardiovascular risk factors in individuals with type 2 diabetes mellitus. Archives of Internal Medicine, 170 1566 1575. T immerman, G. M. (1999). Binge eating scale: further assessment of validity and reliability. Journal of Biobehavioral Research, 4 1 12. Toplak, H., Ziegler, O., Keller, U., Hamann, A., Godin, C., Wittert, G., et al. (2005). X PERT: weight reduction with o rlistat in obese subjects receiving a mildly or moderately reduced energy diet. Early response to treatment predicts weight maintenance. Diabetes, Obesity and Metabolism, 7 699 708. Toubro, S., & Astrup, A. (1997). Randomised comparison of diets for main taining obese fixed energy intake. British Medical Journal, 4 314 329.
107 Trabulsi, J., & Schoeller, D. A. (2001). Evaluation of dietary assessment instruments against doub ly labeled water, a biomarker of habitual energy intake. American Journal of Physiology, Endocrinology, and Metabolism, 281 E891 E899. Tuomilehto, J., Lindstrom, J., Eriksson, J. G., Valle, T. T., Hamalainen, H., Ilanne Parikka, P., et al. (2001). Preve ntion of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. New England Journal of Medicine, 344, 1343 1350. U.S. Department of Health and Human Services. (1996). Physical activity and health: A report of t he Surgeon General. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion. U.S. Department of Health and Human Services and U.S. Department of Agriculture (2005). Dietary g uideline s for Americans, 2005 6 th Ed. Washington, DC: U.S. Government Printing Office. Vazquez, C., Montagna, C., Alcaraz, F., Balsa, J. A., Zamarron, I., Arrieta, F., et al. (2009). Meal replacement with a low calorie diet formula in weight loss maintenance aft er weight loss induction with diet alone. European Journal of Clinical Nutrition 63, 1226 1232 Vetter, M. L., Faulconbridge, L. F., Webb, V. L., & Wadden, T. A. (2010). Behavioral and pharmacologic therapies for obesity. Nature Reviews Endocrinology, 6 578 588. Vgontzas, A. N., Tan, T. L., Bixler, E. O., Martin, L. F., Shubert, D., & Kales, A. (1994). Sleep apnea and sleep disruption in obese patients. Archives of Internal Medicine, 154 1705 1711. Vogels, N., & Westerterp Plantenga, M. S. (2005). Cat egorical strategies based on subject characteristics of dietary restraint and physical activity, for weight maintenance. International Journal of Obesity, 29 849 857. Wadden, T. A., Anderson, D. A., & Foster, G. D. (1999). Two year changes in lipids and lipoproteins associated with the maintenance of a 5% to 10% reduction in initial weight: some findings and some questions. Obesity Research, 7 170 178. Wadden, T. A., Berkowitz, R. J., Sarwer, D. B., Prus Wisniewski, R., Steinberg, C. (2001). Benefits of lifestyle modification in the pharmacologic treatment of obesity. Archives of Internal Medicine, 161 218 227. Wadden, T. A., Berkowitz, R. J, Womble, L. G. Sarwer, D. B., Phelan, S., Cato, R. K., et al. (2005). Randomized trial of lifestyle modification and pharmacotherapy for obesity. New England Journal of Medicine, 353 2111 2120.
108 Wadden, T. A., Butryn, M. L., & Wilson, C. (2007). Lifestyle modification for the management of obesity. Gastroenterology, 132 2226 2238. Wadden, T. A., Foreyt, J. P., Fo ster, G. D., Hill, J. M., et al. (2010). Weight loss with naltrexone SR/buproprion SR combination therapy as an adjunct to behavior modification: The COR BMOD Trial. Obesity, 19 110 120. Wadden, T. A., & Foster, G. D. (1992). B ehavioral assessment and treatment of markedly obese patients. In T. A. Wadden & T. B. Van Itallie (Eds.), Treatment of the s eriously o bese p atient (pp. 290 330). New York NY : Guilford Press. Wadden, T. A., Foster, G. D., & Letizia, K. A. (1994). One year behavioral treatment of obesity: comparison of moderate and severe caloric restriction and the effects of eight maintenance therapy. Journal of Consulting and Clinical Psychology, 62 165 171. Wadden, T. A., Foster, G. D., Letizia, K. A., & Wilk, J. E. (1993). Metabolic, anthropometric, and psychological characteristics of obese binge eaters. International Journal of Eating Disorders, 14 17 25. Wadden, T. A., Foster, G. D., Wang, J., Pierson, R. N., Yang, M. U., Moreland, K., et al. (1992). Clinical correlates of short and long term weight loss. American Journal of Clinical Nutrition, 56 271S 274S. Wadden, T. A., Sternberg, J. A., Letizia, K. A., Stunkard, A. J., & Foster, G. D. (1989). Treatment of obesity by very low calorie diet, behavior thera py, and their combination: a five year perspective. International Journal of Obesity, 13 s39 s46. Wadden, T. A., & Stunkard, A. J. (1986). Controlled trial of very low calorie diet, behavior therapy, and their combination in the treatment of obesity. Jou rnal of Consulting and Clinical Psychology, 54 482 488. Wadden, T. A., Stunkard, A. J., & Brownell, K. D. (1983). Very low calorie diets: Their efficacy, safety, and future. Annals of Internal Medicine, 99 675 684. Wadden, T. A., West, D. S., Neilberg, R. H., Wing, R. R., Ryan, D. H., Johnson, K. C., et al. (2009). One year weight losses in the Look AHEAD study: factors associated with success. Obesity, 17 713 722. Wadden, T. A., Womble, L. G., Stunkard, A. J., & Anderson, D. A. (2002). Psychosocial consequences of obesity and weight loss. In T. A. Wadden and A. J. Stunkard (Eds.), Handbook of o besity t reatment (pp. 144 169). New York NY : Guilford Press.
109 Wardle, J., & Beinart, H. (1981). Binge eating: a theoretical review. British Journal of Clini cal Psychology, 20 97 109. Ware, J. E., Kosinski, M., & Keller, S. K. (1994). SF 36 p hysical and m ental h ealth s ummary s cales: a u m anual Boston, MA: The Health Institute. Warziski, M. T., Sereika, S. M., Styn, M. A., Music, E., & Burke, L. E. (20 08). Changes in self efficacy and dietary adherence: the impact on weight loss in the PREFER study. Journal of Behavioral Medicine, 21 81 92. Weiss, E. C., Galuska, D. A., Kettel Khan, L., Gillespie, C., & Serdula, M. K. (2007). Weight regain in U.S. adu lts who experienced substantial weight loss; 1999 2002. American Journal of Preventative Medicine, 33 34 40. Wilson, P. B., Sullivan, L., Parise, H., & Kannel, W. B. (2002). Overweight and obesity as determinants of cardiovascular risk : the Framingham experience. Archives of Internal Medicine, 162 1867 1872. Wing, R. R. (1998). Behavioral approaches to the treatment of obesity. In G. A. Bray, C. Bouchard, & W. P. T. James (Eds.), Handbook of obesity (pp. 855 873). New York NY : Marce l Dekker. Wing, R. (2002). Behavioral weight control. In T. A. Wadden & A. J. Stunkard (Eds.), Handbook of obesity treatment (pp. 301 316). New York NY : Guilford Press. Wing, R. R., Blair, E., Marcus, M., Epstein, L. H., & Harvey, J. (1994). Year long w eight loss treatment for obese patients with Type II diabetes: Does inclusion of an intermittent very low calorie diet improve outcome? American Journal of Medicine, 97 354 362. Wing, R. R., & Hill, J. O. (2001). Successful weight loss maintenance. Annua l Review of Nutrition, 21 323 341. Wing, R. R., Jeffery, R. W., Burton, L. R., Thorson, C., Nissinoff, K. S., & Baxter, J. E. (1996). Food provision vs structured meal plans in the behavioral treatment of obesity. International Journal of Obesity and Rel ated Metabolic Disorders, 20 56 62. Wing, R. R., Marcus, M. D., Salata, R., Epstein, L. H., Miaskiewicz, S., & Blair, E. H. (1991). Effects of a very low calorie diet on long term glycemic control in obese Type 2 diabetic subjects. Archives of Internal M edicine, 151 1334 1340. Wing, R. R., Papandonatos, G., Fava, J. L., Gorin, A. A., Phelan, S., McCaffery, J., & Tate, D. F. (2008). Maintaining large weight losses: the role of behavioral and psychological factors. Journal of Consulting and Clinical Psych ology, 76 1015 1021.
110 Wing, R. R., Tate, D. F., Gorin, A. A., Raynor, H. A., & Fava, J. L. (2006). A self regulation program for maintenance of weight loss. New England Journal of Medicine, 355 1563 1571.
111 BIOGRAPHICAL SKETCH Lisa M arie Nackers is originally from Wrightstown, Wisconsin and graduated summa cum laude from the University of Wisconsin Eau Claire with a Bachelor of Arts in psychology. Prior to entering graduate school, she accepted a two year Intramural Research Training Award fellowship at the National Institute of Diabetes and Digestive and Kidney Diseases and the National Center for Complementary and Alternative Medicine within the Nat ional Institutes of Health in Bethesda, Maryland Following this research training, she enrolled in the doctorate program in Clinical and Health Psychology at the University of Florida Under the mentorship of Dr. Michael G. Perri, she researched weight loss interventions for obesity, specifically in underserved rural counties She earn ed her Master of Science degree in 2008 in clinical psychology. Concurrently while pursuing her doctor ate, Lisa completed a degree, with an emphasis in social and behavioral sciences, at the University of Florida in December 2010 To fulfill the final requirements of the C linical and Health P sychology program Lisa completed her internship in clinical psychology at Rush Unive rsity Medical Center in Chicago She obtained her D octor of P hilosophy degree in August 2012. Her clinic al interests lie mainly in the area of behavioral medicine. Her research interests center around weight management, lifestyle interventions for obesity, and treatment dissemination.